Airway management method and device for ultrasonic image based on artificial intelligence

Abstract

The invention relates to the technical field of medical information, in particular to an airway management method and device for an ultrasonic image based on artificial intelligence, and the method comprises the steps: obtaining a historical airway ultrasonic image which is manually segmented and marked with the name of each part; inputting the historical airway ultrasound image into the neural network model for training to obtain an airway part identification model; acquiring a target airway ultrasound image of the target patient; based on the identification model of each part of the airway, performing color tagging and anatomical tagging on each part in the target airway ultrasonic image; determining related parameter positions of all the parts based on the coloring marks and the anatomical marks of all the parts in the target airway ultrasound image; based on the relevant parameter positions of all the parts, measuring the parameters of all the parts to obtain parameter measurement values of all the parts; the comprehensive airway risk weighted score of the target patient is determined based on the parameter measurement value of each part, so that the airway ultrasonic evaluation efficiency of a doctor is remarkably improved.

Description

Technical Field

[0001] The present invention relates to the field of medical information technology, and in particular to an airway management method and device based on artificial intelligence ultrasound imaging. Background Art

[0002] In the field of surgical anesthesia, emergency rescue and critical care, keeping the airway open is crucial to maintaining the patient's respiratory function and ensuring the oxygen supply in the patient's body. However, the patient's airway management often faces three risks: 1. Preoperative airway condition assessment; 2. Difficulty or failure of mask ventilation and artificial airway establishment during surgery; 3. Identification of postoperative airway complications.

[0003] The patient's head and neck anatomical factors may cause the spatial structure of the airway to be pulled, squeezed, twisted, and deformed by the surrounding tissues, thereby affecting the patency of the airway.

[0004] Some external visual assessment methods can predict difficult airway, such as facial inspection (Wilson risk score: weight, mandibular mobility, head and neck mobility, mandibular retraction, incisor protrusion), Mallampati classification (grade I: soft palate, uvula, palatal pharyngeal arch and hard palate can be seen; grade II: soft palate, uvula and part of palatal pharyngeal arch can be seen; grade III: only the soft palate and the base of the uvula can be seen; grade IV: only the hard palate can be seen).

[0005] Similarly, laryngoscopy can also be used to perform visual examination of the inside of the oral cavity and pharynx, such as the Cormack-Lehane grading system, which is divided into four grades according to the laryngeal structures that can be seen under laryngoscopy: Grade I: the glottis can be fully exposed; Grade II: the arytenoid cartilages (the posterior wall of the glottis entrance) and the posterior half of the glottis can be seen; Grade III: only the epiglottis can be seen; Grade IV: the epiglottis cannot be seen.

[0006] Although the above methods are used clinically, the sensitivity and specificity of the above airway rating tools are limited because external visual inspection cannot directly observe the anatomical structure of the airway, and laryngoscopy for internal visual inspection of the airway destroys the relative position relationship of the airway anatomy when the patient is at rest or sleeping.

[0007] Clinically, there are many and complex factors that lead to difficult airways. Currently, more than 90% of difficult airways are unexpected difficult airways during routine clinical diagnosis and treatment, which reflects the current lack of pre-identification of difficult airways. Moreover, compared with predictable difficult airways, unexpected difficult airways significantly increase the risk of hypoxia and the urgency of airway management. Decision makers often need to make airway management decisions within minutes. However, most emergency airways lack the guidance of relevant examinations, and their decisions are often made based on experience, which is somewhat blind. In addition, postoperative management of the airway is often neglected, and the treatment of postoperative airway complications is often not timely enough. For example, atelectasis, pulmonary edema, pneumothorax, and pulmonary infection consolidation are usually confirmed by further chest X-ray, CT or magnetic resonance imaging results after the patient shows clinical symptoms such as hypoxemia, cough and sputum, dyspnea, and fever. Postoperative airway ultrasound-guided airway complication treatment is often more efficient, simple to operate, and has no radiation damage. Routine bedside use will help to identify respiratory complications early and improve patient prognosis.

[0008] Currently, there is still no single assessment method that can serve as the gold standard for difficult airway assessment. Clinically, there is an urgent need to develop an intelligent early warning system that can quickly and accurately identify airway management risks. The airway ultrasound graphic report and decision-making system will greatly shorten the airway assessment time and accuracy, improve the efficiency of airway assessment and management, reduce medical accidents, and ensure patient safety.

[0009] High-resolution ultrasound examination has become a recommended method for preoperative airway condition assessment, intraoperative artificial airway establishment, and postoperative airway complication identification because it can accurately reveal the anatomical details of the tissues surrounding the airway. However, ultrasound image acquisition and interpretation require high technical requirements and a complex process. Skilled airway ultrasound standard section scanning and interpretation often requires long-term learning and experience accumulation. This technology is not conducive to surgical anesthesia, emergency rescue, and critical care medical personnel to quickly master.

[0010] Therefore, how to quickly obtain standard airway ultrasound assessment is a technical problem that needs to be solved urgently. Summary of the invention

[0011] In view of the above problems, the present invention provides an airway management method and device based on artificial intelligence ultrasound imaging that overcomes the above problems or at least partially solves the above problems.

[0012] In a first aspect, the present invention provides an airway management method based on ultrasound imaging using artificial intelligence, comprising:

[0013] Acquire an identification model of each part of the airway, wherein the identification model of each part of the airway is obtained by training based on historical airway ultrasound images;

[0014] Acquire an ultrasound image of a target airway of a target patient;

[0015] Based on the airway part recognition model, coloring and anatomically marking each part in the target airway ultrasound image;

[0016] Determining the parameter positions related to each part based on the color annotations and anatomical markers of each part in the target airway ultrasound image;

[0017] Based on the parameter positions related to the parts, the parameters of the parts are measured to obtain the parameter measurement values ​​of the parts;

[0018] Based on the parameter measurements related to each site, a comprehensive airway risk weighted score of the target patient is determined.

[0019] Preferably, after determining the comprehensive airway risk weighted score of the target patient based on the parameter measurement values ​​of each part, the method further includes:

[0020] Based on the comprehensive airway risk weighted score, a difficult airway warning is performed and corresponding airway management decision recommendations are provided.

[0021] Preferably, obtaining the identification model of each part of the airway includes:

[0022] Acquire a historical airway ultrasound image, wherein the historical airway ultrasound image is manually segmented and annotated with the name of each part;

[0023] The historical airway ultrasound images are used as input data, and the corresponding annotated names of the parts are used as output data to be input into a neural network model for training, so as to obtain an airway part recognition model, and the airway part recognition model is verified and optimized.

[0024] Preferably, the names of the parts include:

[0025] The tongue, tongue base, hyoid bone, epiglottis, pre-epiglottic space, vocal cords, glottic symphysis, thyroid cartilage, cricoid cartilage, thyroid gland, trachea, endotracheal tube, esophagus and thoracic cavity.

[0026] Preferably, based on the parameter positions related to the parts, measuring the parameters related to the parts to obtain the parameter measurement values ​​related to the parts includes:

[0027] Based on the relevant parameter positions of each part, the following parameters are measured:

[0028] Tongue width, tongue thickness, tongue volume, distance from skin to epiglottis, distance from skin to vocal cords, distance from skin to hyoid bone, distance from neutral hyoid bone to chin, distance from hyperextended hyoid bone to chin, ratio of the distance from neutral hyoid bone to chin to the distance from hyperextended hyoid bone to chin, distance from skin to tracheal rings of thyroid isthmus;

[0029] Obtain parameter measurement values ​​related to each part.

[0030] Preferably, determining the comprehensive airway risk weighted score of the target patient based on the parameter measurement values ​​related to each part includes:

[0031] Based on the parameter measurements related to each part, the comprehensive airway risk weighted score of the target patient is determined according to the following calculation formula:

[0032]

[0033] Where n is the number of parameter measurements, Δ i is the weight of the parameter measurement, d is the deviation function, j min,i is the lower limit of the normal parameter value of the parameter measurement value, j max,i is the upper limit of the normal parameter value of the parameter measurement value, x i is any parameter measurement value, and S is the comprehensive airway risk weighted score.

[0034] Preferably, the deviation function is defined as follows:

[0035]

[0036] In a second aspect, the present invention further provides an airway management device based on artificial intelligence ultrasound imaging, comprising:

[0037] A first acquisition module is used to acquire historical airway ultrasound images, wherein the historical airway ultrasound images are manually segmented and labeled with names of various parts;

[0038] The first obtaining module is used to input the historical airway ultrasound image into a neural network model for training to obtain an airway part recognition model;

[0039] A second acquisition module is used to acquire a target airway ultrasound image of a target patient;

[0040] A labeling module, used for coloring and annotating various parts in the target airway ultrasound image based on the airway part recognition model;

[0041] A first determination module is used to determine the parameter position related to each part based on the color annotation and anatomical mark of each part in the target airway ultrasound image;

[0042] A second obtaining module is used to measure the parameters of each part based on the parameter positions related to each part to obtain the parameter measurement values ​​of each part;

[0043] The second determination module is used to determine the comprehensive airway risk weighted score of the target patient based on the parameter measurement values ​​of each part.

[0044] In a third aspect, the present invention further provides a computer device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the method described in the first aspect when executing the program.

[0045] In a fourth aspect, the present invention further provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the method described in the first aspect.

[0046] One or more technical solutions in the embodiments of the present invention have at least the following technical effects or advantages:

[0047] The present invention provides an airway management method of ultrasonic imaging using artificial intelligence, comprising: obtaining historical airway ultrasonic images, the historical airway ultrasonic images being manually segmented and labeled with the names of various parts; inputting the historical airway ultrasonic images into a neural network model for training to obtain an identification model for various airway parts; obtaining a target airway ultrasonic image of a target patient; based on the identification model for various airway parts, coloring and annotating various parts in the target airway ultrasonic image and performing anatomical marking; based on the coloring and anatomic marking of various parts in the target airway ultrasonic image, determining the position of parameters related to various parts; based on the position of parameters related to various parts, measuring the parameters of various parts to obtain the parameter measurement values ​​of various parts; based on the parameter measurement values ​​of various parts, determining the comprehensive airway risk weighted score of the target patient, quickly determining the position of parameters related to various parts through the coloring and anatomic marking of the airway ultrasonic images using artificial intelligence, and obtaining the parameter measurement values, thereby quickly obtaining a comprehensive airway risk assessment result, thereby improving the assessment efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] Various other advantages and benefits will become apparent to those of ordinary skill in the art by reading the detailed description of the preferred embodiments below. The accompanying drawings are only for the purpose of illustrating the preferred embodiments and are not to be considered as limiting the present invention. Also, throughout the accompanying drawings, the same reference figures are used to represent the same components. In the drawings:

[0049] Figure 1 A schematic diagram of the steps of the airway management method based on ultrasound imaging using artificial intelligence in an embodiment of the present invention is shown;

[0050] Figure 2A flowchart of an airway management method based on ultrasonic imaging using artificial intelligence in an embodiment of the present invention is shown;

[0051] Figure 3 A schematic diagram of the structure of an airway management device based on ultrasonic imaging of artificial intelligence in an embodiment of the present invention is shown;

[0052] Figure 4 A schematic diagram of the structure of a computer device for implementing an airway management method of ultrasonic imaging based on artificial intelligence in an embodiment of the present invention is shown. DETAILED DESCRIPTION

[0053] Exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although exemplary embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be limited by the embodiments set forth herein. On the contrary, these embodiments are provided to enable a more thorough understanding of the present disclosure and to fully convey the scope of the present disclosure to those skilled in the art.

[0054] Embodiment 1

[0055] The embodiment of the present invention provides an airway management method based on ultrasound imaging of artificial intelligence, such as Figure 1 As shown, including:

[0056] S101, obtaining an identification model of each part of the airway, where the identification model of each part of the airway is trained based on historical airway ultrasound images;

[0057] S102, acquiring a target airway ultrasound image of a target patient;

[0058] S103, based on the airway part recognition model, coloring and anatomy marking each part in the target airway ultrasound image;

[0059] S104, determining the parameter position related to each part based on the color annotation and anatomical mark of each part in the target airway ultrasound image;

[0060] S105, based on the parameter positions related to each part, measuring the parameters of each part to obtain the parameter measurement values ​​of each part;

[0061] S106, determining a comprehensive airway risk weighted score for the target patient based on the parameter measurement values ​​related to each part.

[0062] First, if Figure 2 As shown, the overall scheme of the present invention is described:

[0063] By collecting the target airway ultrasound image of the target patient, the position and name of each part in the target airway ultrasound image are identified through the airway part recognition model, the parameters related to each part are measured, and the parameter measurement values ​​related to each part are obtained. Through evaluation, the comprehensive airway risk weighted score of the target patient is obtained, and according to the comprehensive airway risk weighted score, corresponding airway management decision recommendations are provided. Finally, according to the comprehensive airway risk weighted score and airway management decision recommendations, a graphic report is generated for the doctor's reference.

[0064] The AI-based ultrasound image segmentation technology improves doctors’ ability to identify and judge standard airway ultrasound images, thereby lowering the technical threshold for airway ultrasound section scanning, interpretation and decision-making, and reducing the workload of airway ultrasound. The following is a detailed introduction to each step:

[0065] In an optional implementation, S101 includes:

[0066] Obtain historical airway ultrasound images, which are manually segmented and labeled with the names of various parts; use the historical airway ultrasound images as input data, and input the corresponding labeled names of various parts as output data into a neural network model for training to obtain an airway part recognition model, and verify and optimize the airway part recognition model.

[0067] Specifically, the historical airway ultrasound images may be airway ultrasound images provided by many volunteers, which are manually segmented and labeled with the names of various parts by a team of experts in the airway field, including tongue body, tongue root, hyoid bone, epiglottis, pre-epiglottic space, vocal cords, glottic symphysis, thyroid cartilage, cricoid cartilage, thyroid gland, trachea, endotracheal tube, esophagus, and chest cavity.

[0068] The historical airway ultrasound images that have been manually segmented and labeled with the names of various parts are input into the neural network model for training to obtain an airway part recognition model. The airway part recognition model can label the airway parts of any airway ultrasound image when any airway ultrasound image is input.

[0069] Of course, after obtaining the initial airway part identification model, it is continuously optimized through verification, expansion of the amount of training data, optimization of algorithms, and airway ultrasound expert judgment to obtain a more effective airway part identification model.

[0070] Next, S102 is executed to obtain a target airway ultrasound image of the target patient, which is obtained by connecting to an external ultrasound machine through an ultrasound image receiving port, where the external connection interface may be USB, Type-C, VGA, HDMI, DVI, etc.

[0071] S103, based on the airway part recognition model, color and annotate each part in the target airway ultrasound image.

[0072] Specifically, the target airway ultrasound image is input into the airway part recognition model, and the output result is the name of each part and the corresponding part position. Then, each part is color-labeled and anatomically marked. That is, different parts are distinguished and labeled by different colors, and different parts are marked with corresponding names.

[0073] Next, S104 is executed to determine the parameter position related to each part based on the color annotation and anatomical mark of each part in the target airway ultrasound image.

[0074] In the target airway ultrasound image, each part is marked with color, and the parameters related to each part need to be determined by professionals based on the color marking. The parameters related to each part here are specifically the distance between each part and other parameters. After the position of each part is determined, the distance between any two parts can be determined.

[0075] During ultrasonic image recognition, the measured values ​​of these parameters can be clearly marked. Therefore, S105 is executed to measure the parameters of each part based on the parameter positions related to each part to obtain the measured values ​​of the parameters of each part.

[0076] In the present invention, the evaluation criteria of airway difficulty can be measured in the following manner to obtain the parameter measurement values ​​related to each part: wherein airway difficulty generally refers to the difficulty of mask ventilation and endotracheal intubation under direct laryngoscopy.

[0077] Specifically, based on the parameter positions related to each part, the parameters of the following parts are measured:

[0078] Tongue width, tongue thickness, tongue volume, distance from skin to epiglottis, distance from skin to vocal cords, distance from skin to hyoid bone, distance from neutral hyoid bone to chin, distance from hyperextended hyoid bone to chin, ratio of distance from neutral hyoid bone to chin to distance from hyperextended hyoid bone to chin, distance from skin to tracheal rings of thyroid isthmus;

[0079] Obtain the parameter measurement values ​​related to each part.

[0080] After obtaining the parameter measurement values ​​related to each part, S106 is executed to determine the comprehensive airway risk weighted score of the target patient based on the parameter measurement values ​​related to each part.

[0081] Specifically, the comprehensive airway risk weighted score of the target patient is determined according to the following calculation formula:

[0082]

[0083] Where n is the number of parameter measurements, Δ i is the weight of any parameter measurement, d is the deviation function, j min,i is the lower limit of the normal parameter value of any parameter measurement, j max,i is the upper limit of the normal parameter value of any parameter measurement, x i is the measured value of any parameter, and S is the comprehensive airway risk weighted score.

[0084] The deviation function is defined as follows:

[0085]

[0086] Among them, the upper and lower limits of the normal parameter value of any parameter measurement value [j min,i , j max,i ] and the weight of any parameter measurement value are obtained from evidence-based medicine such as the latest research or guidelines at home and abroad. The comprehensive airway risk weighted score is obtained according to the above formula.

[0087] After obtaining the comprehensive airway risk weighted score, it also includes:

[0088] Based on the comprehensive airway risk weighted score, difficult airway warning is performed and corresponding airway management decision recommendations are provided.

[0089] When the comprehensive airway risk weighted score is high, a difficult airway warning is performed. The comprehensive airway risk weighted score can be divided into different levels, which are not limited here.

[0090] A corresponding relationship table of different comprehensive airway risk weighted scores and corresponding airway management decision suggestions is stored in the database. By looking up the table, the corresponding airway management decision suggestion can be determined and output through the display screen.

[0091] Specifically, for airway management, the corresponding decision-making recommendations are:

[0092] Standardized airway management:

[0093] 1. Assess the airway

[0094] Before anesthesia or airway management is initiated, an airway risk assessment should be performed by the person responsible for airway management, when feasible, to identify patient, medical, surgical, environmental, and anesthetic factors (e.g., risk of aspiration) that may indicate a potentially difficult airway.

[0095] Before anesthesia or airway management is initiated, a physical examination of the airway is performed to further identify physical features that may indicate a potential difficult airway.

[0096] 2. Preparation for Difficult Airway Management

[0097] Ensure airway management equipment is available in the room.

[0098] Ensure that portable storage units containing specialized equipment for difficult airway management are readily available.

[0099] If a difficult airway is known or suspected, ensure a skilled individual is present or immediately available to assist with airway management, if feasible.

[0100] Inform the patient or responsible person about the special risks and procedures of difficult airway management.

[0101] Administer supplemental oxygen before initiating difficult airway management and continue supplemental oxygen as much as possible during difficult airway management, including extubation.

[0102] Ensure that, at a minimum, the ASA Basic Anesthesia Monitoring Standards are followed to monitor all patients before, during, and after airway management.

[0103] 3. Anticipated Difficult Airway Management

[0104] Develop a pre-planned strategy for anticipated difficult airway management, including awake intubation, patients who can be adequately ventilated but are difficult to intubate, patients who are unable to be ventilated or intubated, and difficulties with emergency invasive airway rescue.

[0105] When appropriate, perform awake intubation if difficult intubation is suspected and ventilation (mask / supraglottic airway) is anticipated to be difficult, the risk of aspiration is increased, the patient may not tolerate brief apneas, or urgent invasive airway rescue is anticipated to be difficult.

[0106] 4. Unexpected and Emergency Difficult Airway Management

[0107] Call for help.

[0108] Optimize oxygenation.

[0109] Determine the benefits of awakening and / or restoring spontaneous breathing.

[0110] To determine the benefits of non-invasive versus invasive airway management methods.

[0111] If a non-invasive approach is chosen, prioritize the non-invasive devices to be used.

[0112] 5. Confirm endotracheal intubation

[0113] Use capnography or end-tidal carbon dioxide monitoring to confirm endotracheal intubation.

[0114] 6. Extubation of difficult airway

[0115] Develop a pre-planned strategy for extubation and subsequent airway management.

[0116] 7. Follow-up care

[0117] Use postextubation steroids and / or epinephrine when appropriate.

[0118] Inform the patient or responsible person of airway difficulties encountered to guide and facilitate the provision of future care.

[0119] Document the nature and quality of the airway difficulty in the medical record to guide and facilitate the provision of future care.

[0120] Instruct patients to register for emergency notification services, where appropriate and feasible.

[0121] Emergency airway management:

[0122] 1. Ensure ventilation and oxygenation, and at the same time, preliminarily assess the airway condition according to the "CHANNEL principle", with the primary goal of ensuring the patient's life safety.

[0123] 2. Clarify the airway situation and establish an artificial airway. When encountering a difficult airway, follow the principle of "maintaining ventilation and oxygenation first" and avoid blindly trying multiple times. The way to establish an artificial airway follows the principle of "simple, effective, and minimally invasive", and visualization technology is preferred.

[0124] The following airway management decision suggestions can also be made:

[0125] New non-invasive ventilation technology, anterior neck invasive airway technology and multidisciplinary collaboration methods, etc.

[0126] Finally, a graphic report can be generated, which includes: target patient information, target airway ultrasound image, report information content, comprehensive airway risk score and corresponding airway management decision recommendations.

[0127] The technical solution of the present invention can significantly improve the efficiency of doctors in airway ultrasound assessment, reduce the errors and workload of manual assessment, and reduce the risk of airway complications by automatically identifying airway risks and quickly and accurately obtaining recommended airway management methods. At the same time, automatically generated graphic reports and airway management decisions will help doctors view and make a series of key and effective airway management decisions, thereby reducing the incidence of unexpected difficult airways, reducing the difficulty or risk of failure of mask ventilation and manual airway establishment, and timely identifying postoperative airway complications. Ensure the clinical safety of patients.

[0128] One or more technical solutions in the embodiments of the present invention have at least the following technical effects or advantages:

[0129] The present invention provides an airway management method for ultrasonic imaging based on artificial intelligence, comprising: obtaining historical airway ultrasonic images, which are manually segmented and labeled with the names of various parts; inputting the historical airway ultrasonic images into a neural network model for training to obtain an identification model for various airway parts; obtaining a target airway ultrasonic image of a target patient; based on the identification model for various airway parts, coloring and annotating various parts in the target airway ultrasonic image and performing anatomical marking; based on the coloring and anatomic marking of various parts in the target airway ultrasonic image, determining the position of parameters related to each part; based on the position of parameters related to each part, measuring the parameters of each part to obtain the parameter measurement values ​​of each part; based on the parameter measurement values ​​of each part, determining the comprehensive airway risk weighted score of the target patient, and quickly determining the position of parameters related to each part and obtaining the parameter measurement values ​​by coloring and annotating the airway ultrasonic image with artificial intelligence, and then quickly obtaining a comprehensive airway risk assessment result, thereby improving the assessment efficiency.

[0130] Embodiment 2

[0131] Based on the same inventive concept, the embodiment of the present invention also provides an airway management device based on artificial intelligence ultrasound imaging, such as Figure 3 As shown, including:

[0132] The first acquisition module 301 is used to acquire historical airway ultrasound images, where the historical airway ultrasound images are manually segmented and labeled with names of various parts;

[0133] The first obtaining module 302 is used to input the historical airway ultrasound image into a neural network model for training to obtain an airway part recognition model;

[0134] A second acquisition module 303 is used to acquire a target airway ultrasound image of a target patient;

[0135] A labeling module, used for coloring and annotating various parts in the target airway ultrasound image based on the airway part recognition model;

[0136] A first determination module 304 is used to determine the parameter position related to each part based on the color annotation and anatomical mark of each part in the target airway ultrasound image;

[0137] The second obtaining module 305 is used to measure the parameters of the parts based on the parameter positions related to the parts to obtain the parameter measurement values ​​of the parts;

[0138] The second determination module 306 is used to determine the comprehensive airway risk weighted score of the target patient based on the parameter measurement values ​​of each part.

[0139] In an optional implementation, the system further includes an early warning module, configured to:

[0140] Based on the comprehensive airway risk weighted score, a difficult airway warning is performed and corresponding airway management decision recommendations are provided.

[0141] In an optional implementation, the first acquisition module 301 is used to:

[0142] Acquire a historical airway ultrasound image, wherein the historical airway ultrasound image is manually segmented and annotated with the name of each part;

[0143] The historical airway ultrasound images are used as input data, and the corresponding annotated names of the parts are used as output data to be input into a neural network model for training, so as to obtain an airway part recognition model, and the airway part recognition model is verified and optimized.

[0144] In an optional embodiment, the names of the parts include:

[0145] The tongue, tongue base, hyoid bone, epiglottis, pre-epiglottic space, vocal cords, glottic symphysis, thyroid cartilage, cricoid cartilage, thyroid gland, trachea, endotracheal tube, esophagus and thoracic cavity.

[0146] In an optional implementation, the second obtaining module 305 is used to:

[0147] Based on the relevant parameter positions of each part, the following parameters are measured:

[0148] Tongue width, tongue thickness, tongue volume, distance from skin to epiglottis, distance from skin to vocal cords, distance from skin to hyoid bone, distance from neutral hyoid bone to chin, distance from hyperextended hyoid bone to chin, ratio of the distance from neutral hyoid bone to chin to the distance from hyperextended hyoid bone to chin, distance from skin to tracheal rings of thyroid isthmus;

[0149] Obtain parameter measurement values ​​related to each part.

[0150] In an optional implementation, the second determining module 306 is configured to:

[0151] Based on the parameter measurements related to each part, the comprehensive airway risk weighted score of the target patient is determined according to the following calculation formula:

[0152]

[0153] Where n is the number of parameter measurements, Δ i is the weight of the parameter measurement, d is the deviation function, j min,i is the lower limit of the normal parameter value of the parameter measurement value, j max,i is the upper limit of the normal parameter value of the parameter measurement value, x iis any parameter measurement value, and S is the comprehensive airway risk weighted score.

[0154] In an optional implementation, the deviation function is defined as follows:

[0155]

[0156] Embodiment 3

[0157] Based on the same inventive concept, an embodiment of the present invention provides a computer device, such as Figure 4 As shown, it includes a memory 404, a processor 402, and a computer program stored in the memory 404 and executable on the processor 402. When the processor 402 executes the program, the steps of the above-mentioned artificial intelligence-based ultrasound imaging airway management method are implemented.

[0158] Among them, Figure 4 In the embodiment of the present invention, a bus architecture (represented by bus 400) is shown, which may include any number of interconnected buses and bridges, and bus 400 links various circuits including one or more processors represented by processor 402 and memory represented by memory 404. Bus 400 may also link various other circuits such as peripherals, voltage regulators, and power management circuits, which are well known in the art and are therefore not further described herein. Bus interface 406 provides an interface between bus 400 and receiver 401 and transmitter 403. Receiver 401 and transmitter 403 may be the same element, namely a transceiver, providing a unit for communicating with various other devices over a transmission medium. Processor 402 is responsible for managing bus 400 and general processing, while memory 404 may be used to store data used by processor 402 when performing operations.

[0159] Embodiment 4

[0160] Based on the same inventive concept, an embodiment of the present invention provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the above-mentioned artificial intelligence-based ultrasound imaging airway management method.

[0161] The algorithm and display provided herein are not inherently related to any particular computer, virtual system or other device. Various general purpose systems can also be used together with the teachings based on this. According to the above description, it is obvious that the structure required for constructing such systems. In addition, the present invention is not directed to any specific programming language either. It should be understood that various programming languages ​​can be utilized to realize the content of the present invention described herein, and the description of the above specific languages ​​is for disclosing the best mode of the present invention.

[0162] In the description provided herein, a large number of specific details are described. However, it is understood that embodiments of the present invention can be practiced without these specific details. In some instances, well-known methods, structures and techniques are not shown in detail so as not to obscure the understanding of this description.

[0163] Similarly, it should be understood that in order to streamline the present disclosure and aid in understanding one or more of the various inventive aspects, in the above description of exemplary embodiments of the present invention, the various features of the present invention are sometimes grouped together into a single embodiment, figure, or description thereof. However, this disclosed method should not be interpreted as reflecting the following intention: that the claimed invention requires more features than the features explicitly recorded in each embodiment. More specifically, as reflected in each embodiment, the inventive aspects lie in less than all the features of the single embodiment disclosed above. Therefore, the claims that follow the specific embodiment are hereby expressly incorporated into the specific embodiment, with each claim itself serving as a separate embodiment of the present invention.

[0164] Those skilled in the art will appreciate that the modules in the devices in the embodiments may be adaptively changed and arranged in one or more devices different from the embodiments. The modules or units or components in the embodiments may be combined into one module or unit or component, and in addition they may be divided into a plurality of submodules or subunits or subcomponents. Except that at least some of such features and / or processes or units are mutually exclusive, all features disclosed in this specification (including the accompanying claims, abstracts and drawings) and all processes or units of any method or device disclosed in this manner may be combined in any combination. Unless otherwise expressly stated, each feature disclosed in this specification (including the accompanying claims, abstracts and drawings) may be replaced by an alternative feature providing the same, equivalent or similar purpose.

[0165] In addition, those skilled in the art will appreciate that, although some embodiments herein include certain features included in other embodiments but not other features, the combination of features of different embodiments is meant to be within the scope of the present invention and form different embodiments. For example, in a specific embodiment, any one of the claimed embodiments may be used in any combination.

[0166] The various component embodiments of the present invention may be implemented in hardware, or in software modules running on one or more processors, or in a combination thereof. Those skilled in the art should understand that a microprocessor or digital signal processor (DSP) may be used in practice to implement some or all of the functions of some or all of the components of an airway management device or computer device based on artificial intelligence ultrasound imaging according to an embodiment of the present invention. The present invention may also be implemented as a device or device program (e.g., a computer program and a computer program product) for executing part or all of the methods described herein. Such a program for implementing the present invention may be stored on a computer-readable medium, or may be in the form of one or more signals. Such a signal may be downloaded from an Internet website, or provided on a carrier signal, or provided in any other form.

[0167] It should be noted that the above embodiments illustrate the present invention rather than limit it, and that those skilled in the art may devise alternative embodiments without departing from the scope of the appended claims. In the claims, any reference symbol between brackets shall not be construed as a limitation on the claims. The word "comprising" does not exclude the presence of elements or steps not listed in the claims. The word "one" or "an" preceding an element does not exclude the presence of a plurality of such elements. The present invention may be implemented by means of hardware comprising a number of different elements and by means of a suitably programmed computer. In a unit claim enumerating a number of devices, several of these devices may be embodied by the same hardware item. The use of the words first, second, and third, etc., does not indicate any order. These words may be interpreted as names.

Claims

1. An airway management method based on ultrasonic imaging using artificial intelligence, characterized in that: include: Acquire an identification model of each part of the airway, wherein the identification model of each part of the airway is obtained by training based on historical airway ultrasound images; Acquire an ultrasound image of a target airway of a target patient; Based on the airway part recognition model, coloring and anatomically marking each part in the target airway ultrasound image; Determining the parameter positions related to each part based on the color annotations and anatomical markers of each part in the target airway ultrasound image; Based on the parameter positions related to the parts, the parameters of the parts are measured to obtain the parameter measurement values ​​of the parts; Based on the parameter measurements related to each site, a comprehensive airway risk weighted score of the target patient is determined.

2. The method according to claim 1, characterized in that After determining the comprehensive airway risk weighted score of the target patient based on the parameter measurement values ​​of each part, the method further includes: Based on the comprehensive airway risk weighted score, a difficult airway warning is performed and corresponding airway management decision recommendations are provided.

3. The method according to claim 1, characterized in that Obtain the identification model of each part of the airway, including: Acquire a historical airway ultrasound image, wherein the historical airway ultrasound image is manually segmented and annotated with the name of each part; The historical airway ultrasound images are used as input data, and the corresponding annotated names of the parts are used as output data to be input into a neural network model for training, so as to obtain an airway part recognition model, and the airway part recognition model is verified and optimized.

4. The method according to claim 3, characterized in that The names of the parts include: The tongue, tongue base, hyoid bone, epiglottis, pre-epiglottic space, vocal cords, glottic symphysis, thyroid cartilage, cricoid cartilage, thyroid gland, trachea, endotracheal tube, esophagus and thoracic cavity.

5. The method according to claim 1, characterized in that Based on the parameter positions related to the parts, the parameters related to the parts are measured to obtain the parameter measurement values ​​related to the parts, including: Based on the relevant parameter positions of each part, the following parameters are measured: Tongue width, tongue thickness, tongue volume, distance from skin to epiglottis, distance from skin to vocal cords, distance from skin to hyoid bone, distance from neutral hyoid bone to chin, distance from hyperextended hyoid bone to chin, ratio of the distance from neutral hyoid bone to chin to the distance from hyperextended hyoid bone to chin, distance from skin to tracheal rings of thyroid isthmus; Obtain parameter measurement values ​​related to each part.

6. The method according to claim 1, characterized in that Based on the parameter measurements related to each part, a comprehensive airway risk weighted score of the target patient is determined, including: Based on the parameter measurements related to each part, the comprehensive airway risk weighted score of the target patient is determined according to the following calculation formula: Where n is the number of parameter measurements, Δ i is the weight of the parameter measurement, d is the deviation function, j min,i is the lower limit of the normal parameter value of the parameter measurement, j max,i is the upper limit of the normal parameter value of the parameter measurement value, x i is the measurement value of any parameter, and S is the comprehensive airway risk weighted score.

7. The method according to claim 6, characterized in that The deviation function is defined as follows:

8. An airway management device based on artificial intelligence ultrasound imaging, characterized in that: include: A first acquisition module is used to acquire historical airway ultrasound images, wherein the historical airway ultrasound images are manually segmented and labeled with names of various parts; The first obtaining module is used to input the historical airway ultrasound image into a neural network model for training to obtain an airway part recognition model; A second acquisition module is used to acquire a target airway ultrasound image of a target patient; A labeling module, used for coloring and annotating various parts in the target airway ultrasound image based on the airway part recognition model; A first determination module is used to determine the parameter position related to each part based on the color annotation and anatomical mark of each part in the target airway ultrasound image; A second obtaining module is used to measure the parameters of each part based on the parameter positions related to each part to obtain the parameter measurement values ​​of each part; The second determination module is used to determine the comprehensive airway risk weighted score of the target patient based on the parameter measurement values ​​of each part.

9. A computer device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the program, the method according to any one of claims 1 to 7 is implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the method according to any one of claims 1 to 7 is implemented.

Images