A endotracheal intubation system, method and electronic device

Abstract

This invention provides an endotracheal intubation system, method, and electronic device, relating to the field of clinical endotracheal intubation technology. It utilizes a trained artificial intelligence model to identify the trachea and esophagus and provides an alarm for incorrect intubation placement, reducing the risk of operational errors by medical personnel. The system includes: an image acquisition module for acquiring laryngoscopy image data; a tagging module for labeling target tracheal images based on the laryngoscopy image data; an image processing module for processing the target tracheal images and simulating specific clinical scenarios; and a model building module for constructing an artificial intelligence model based on the processed target tracheal images, identifying the trachea and esophagus using the artificial intelligence model, and providing an alarm for incorrect intubation placement. The endotracheal intubation system is applied to the aforementioned endotracheal intubation method. The endotracheal intubation method is applied to the electronic device.

Description

Technical Field

[0001] This invention relates to the field of clinical endotracheal intubation technology, and more specifically to an endotracheal intubation system, method, electronic device, and computer-readable storage medium. Background Technology

[0002] Endotracheal intubation is one of the most common procedures in emergency department clinical medical and nursing work, and a crucial life-saving treatment measure. Clinically, endotracheal intubation is often necessary to establish an artificial airway for critically ill patients, especially those with respiratory arrest or those under general anesthesia, to assist their breathing. The procedure is often urgent and has a low margin for error. Due to the physiological and anatomical proximity of the trachea and esophagus, the most common mistake is inserting the tube into the esophagus instead of the trachea. This can directly endanger the patient's life and even lead to death.

[0003] Currently, clinical differentiation of the trachea and esophagus under laryngoscopy mainly relies on the personal experience of medical staff. Using images from a video endoscope, the trachea and esophagus are manually identified and corresponding intubation is performed, but there is a lack of guidance during intubation and warnings of intubation errors. After intubation, common methods for checking the position of the gastric tube include aspirating gastric fluid, listening for gurgling sounds, and observing for air at the external opening of the gastric tube. The position of the endotracheal tube is checked using X-ray machines or carbon dioxide detectors. The detection and judgment of intubation errors takes time, and by the time they are discovered, the consequences have often already occurred.

[0004] Existing patents mostly involve modifying the design of endotracheal tubes and gastric tubes to place detection devices on top. However, such detection can only check whether the tube is correctly positioned after intubation and lacks the function of actively guiding medical staff during intubation. Furthermore, adding detection devices to endotracheal tubes and gastric tubes will inevitably increase the volume of the tubes, and if corresponding detection operations are to be performed in the already narrow oral and pharyngeal space of the patient, it will inevitably increase the difficulty of use for medical staff. Summary of the Invention

[0005] To overcome the problems in existing technologies where the lack of intubation guidance and error prompts can lead to misoperation when manually identifying the trachea and esophagus and performing corresponding intubation procedures, this invention provides a tracheal intubation system, method, electronic device, and computer-readable storage medium.

[0006] This invention provides an endotracheal intubation system, the system comprising:

[0007] Image acquisition module, used to acquire laryngoscope image data;

[0008] The tagging module is used to annotate the target tracheal image based on the laryngoscope image data;

[0009] The image processing module is used to process the target tracheal image and simulate special clinical scenarios, including: foreign bodies or tumors in the pharynx, hemoptysis in the larynx, and anatomical deformities of the pharynx.

[0010] The model building module is used to build an artificial intelligence model based on the processed target trachea image, identify the trachea and esophagus based on the artificial intelligence model, and issue an alarm for incorrect intubation position.

[0011] Preferably, the tag module includes:

[0012] The conversion unit is used to convert the image information of the laryngoscope image data into spectral information based on Fourier transform;

[0013] A labeling assistance unit is used to train a pixel-level labeling assistance component based on a neural network according to the spectral information.

[0014] The selection unit is used to select the target tracheal image based on the annotation auxiliary component;

[0015] The border annotation unit is used to automatically segment the border details of the target trachea image based on image signal processing methods.

[0016] Preferably, the image processing module includes:

[0017] The processing unit is used to process the target tracheal image based on a data augmentation method;

[0018] The simulation unit is used to simulate the specific clinical scenario based on the generative adversarial network method.

[0019] Preferably, the model building module includes:

[0020] The model training unit is used to input the processed target tracheal image into the artificial intelligence model for training, and output the trained artificial intelligence model.

[0021] The model recognition unit is used to identify the trachea and esophagus based on the trained artificial intelligence model and to issue an alarm for incorrect intubation position.

[0022] Preferably, the model training unit is used to input the processed target tracheal image into the artificial intelligence model for training, including:

[0023] The model training unit is used to input the processed target tracheal image into the artificial intelligence model and train the artificial intelligence model using the maximum likelihood function as the loss function.

[0024] Compared with the prior art, the endotracheal intubation system provided by the present invention has the following beneficial effects:

[0025] This invention provides an endotracheal intubation system that utilizes an image acquisition module to acquire laryngoscopy image data, a tagging module to label the laryngoscopy image data to identify the target tracheal image, and an image processing module to process the target tracheal image and simulate special clinical scenarios, such as foreign bodies or tumors in the pharynx, hemoptysis in the larynx, and anatomical deformities of the pharynx. An artificial intelligence model is constructed using the processed target tracheal image. Based on the constructed AI model, the system identifies the location of the trachea and esophagus, providing clear voice and video guidance to medical staff, reducing the risk of operational errors. Furthermore, an alarm is triggered promptly when the intubation position is incorrect, prompting medical staff to make corrections, thereby reducing losses caused by medical operational errors and saving medical staff's time and energy.

[0026] The present invention also provides a method for endotracheal intubation, the method comprising:

[0027] Step 1: Acquire laryngoscope image data;

[0028] Step 2: Annotate the target trachea image based on the laryngoscope image data;

[0029] Step 3: Process the target tracheal image and simulate a special clinical scenario;

[0030] Step 4: Construct an artificial intelligence model based on the processed target trachea image, identify the trachea and esophagus based on the artificial intelligence model, and issue an alarm for incorrect intubation position.

[0031] Preferably, step 2: annotating the target tracheal image based on the laryngoscopy image data includes:

[0032] Step 2.1: Convert the image information of the laryngoscope image data into spectral information based on Fourier transform;

[0033] Step 2.2: Based on the spectral information, train a pixel-level annotation auxiliary component using a neural network;

[0034] Step 2.3: Select the target tracheal image based on the annotation auxiliary component;

[0035] Step 2.4: Automatically segment the border details of the target trachea image based on image signal processing methods;

[0036] Step 3: Processing the target tracheal image and simulating specific clinical scenarios includes:

[0037] Step 3.1: Process the target tracheal image using data augmentation methods;

[0038] Step 3.2: Simulate the specific clinical scenario based on the generative adversarial network method.

[0039] Preferably, step 4: constructing an artificial intelligence model based on the processed target tracheal image, identifying the trachea and esophagus based on the artificial intelligence model, and issuing an alarm for incorrect intubation position, including:

[0040] Step 4.1: Input the processed target tracheal image into the artificial intelligence model, train the artificial intelligence model using the maximum likelihood function as the loss function, and output the trained artificial intelligence model;

[0041] Step 4.2: Identify the trachea and esophagus based on the trained artificial intelligence model, and issue an alarm for incorrect intubation position.

[0042] Compared with the prior art, the beneficial effects of the endotracheal intubation method provided by the present invention are the same as those of the endotracheal intubation system described in the above technical solution, and will not be repeated here.

[0043] The present invention also provides an electronic device, including a bus, a transceiver (display unit / output unit, input unit), a memory, a processor, and a computer program stored in the memory and executable on the processor. The transceiver, the memory, and the processor are connected via the bus. When the computer program is executed by the processor, it implements the steps of any of the above-described endotracheal intubation methods.

[0044] Compared with the prior art, the beneficial effects of the electronic device provided by the present invention are the same as those of the endotracheal intubation system described in the above technical solution, and will not be repeated here.

[0045] The present invention also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of any of the above-described endotracheal intubation methods.

[0046] Compared with the prior art, the beneficial effects of the computer-readable storage medium provided by the present invention are the same as the beneficial effects of the endotracheal intubation system described in the above technical solution, and will not be repeated here.

[0047] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, preferred embodiments are described below in detail with reference to the accompanying drawings. Attached Figure Description

[0048] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0049] Figure 1 A schematic diagram of an endotracheal intubation system provided in an embodiment of the present invention is shown;

[0050] Figure 2 A flowchart of an endotracheal intubation method provided by an embodiment of the present invention is shown;

[0051] Figure 3 This diagram shows an overall framework of an endotracheal intubation method provided by an embodiment of the present invention.

[0052] Figure 4 A schematic diagram of an electronic device for performing a tracheal intubation method provided by an embodiment of the present invention is shown. Detailed Implementation

[0053] In the description of this invention, it should be understood that the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Therefore, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of this invention, "a plurality of" means two or more, unless otherwise explicitly specified.

[0054] In this embodiment, "multiple" refers to two or more. "And / or" describes the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A alone, A and B simultaneously, or B alone. Words such as "exemplary" or "for example" are used to indicate examples, illustrations, or explanations, intended to present related concepts in a specific manner, and should not be construed as superior or more advantageous than other embodiments or designs.

[0055] Because the trachea and esophagus are located close to each other in physiological and anatomical structures, medical staff, especially those with limited intubation experience, are quite likely to confuse the two in high-pressure intensive care settings, leading to operational errors that delay patient rescue and cause serious harm. With the increasing number of intubations, such operational errors not only waste the time and energy of medical staff but can also have fatal consequences for patients.

[0056] In clinical settings, this is typically done in conjunction with a visualization endoscope, which includes all visualization endotracheal intubation equipment, such as a video laryngoscope, a flexible video laryngoscope, and a bronchoscope. The visualization endoscope displays an image of the larynx, and medical staff then manually distinguish the trachea and esophagus to perform the subsequent intubation procedure.

[0057] Existing patents, for example, include: Application No. CN201510674352x, a tracheoesophageal dual-intubation video laryngoscope, which ensures simultaneous tracheoesophageal and esophageal intubation, shortening intubation and rescue time. The main body of the dual-intubation laryngoscope includes an operating handle, a rigid dual-intubation laryngoscope blade, an optical path and camera channel, a camera, a built-in LED light source, and a dual-intubation track. Application No. CN202020275512x, a novel auxiliary detection device for nasogastric tube intubation, has pH test paper or an acid-base detection coating at the front end of the guidewire. After the gastric tube is inserted into the stomach, the presence of the nasogastric tube is detected by observing the color change of the pH test paper or acid-base detection coating. Application number CN201520017563 describes a gastric tube with a detection function. The design involves guiding light through a convex lens tube into an optical fiber, from which the light exits through a notch at the fold. The refracted light is then amplified and dispersed by a concave lens tube, allowing medical personnel to determine if the gastric tube is inserted correctly.

[0058] Existing patents mostly involve modifying the design of endotracheal tubes and gastric tubes to place detection devices on top. However, such detection can only check whether the tube is correctly positioned after intubation and lacks the function of actively guiding medical staff during intubation. Furthermore, adding detection devices to endotracheal tubes and gastric tubes will inevitably increase the volume of the tubes, and if corresponding detection operations are to be performed in the already narrow oral and pharyngeal space of patients, it will inevitably increase the difficulty of use for medical staff.

[0059] Current clinical solutions lack guidance during intubation, relying solely on the experience of medical staff to identify the trachea and esophagus. Post-intubation, proactive error detection by medical staff is essential, and tracheal position detection relies on external facilities such as X-ray machines or carbon dioxide detectors. There is a lack of a system that proactively detects errors and issues warnings.

[0060] Based on this, embodiments of the present invention provide an endotracheal intubation system. Figure 1 A schematic diagram of an endotracheal intubation system provided by an embodiment of the present invention is shown.

[0061] like Figure 1 As shown, the endotracheal intubation system includes: an image acquisition module 1, used to acquire laryngoscope image data.

[0062] It should be noted that, in order to allow the laryngoscope blade to be inserted into the larynx, the image acquisition module 1 can use a small device, such as a miniature camera 400 or a pinhole camera, with its size determined by its ability to be easily inserted into the patient's larynx without causing trauma. The camera can have its own built-in light source or can be illuminated by an external light source. Specifically, data acquisition equipment can be used, or existing equipment in various hospitals can be used directly.

[0063] like Figure 1 As shown, the endotracheal intubation system also includes a tag module 2, which is used to annotate the target tracheal image based on laryngoscope image data.

[0064] like Figure 1 As shown, the labeling module 2 includes: a conversion unit 21, used to convert the image information of the laryngoscope image data into spectral information based on Fourier transform; a labeling auxiliary unit 22, used to train a pixel-level labeling auxiliary component 220 based on the spectral information and a neural network; a selection unit 23, used to select the target tracheal image based on the labeling auxiliary component 220; and a border labeling unit 24, used to automatically segment the border details of the target tracheal image based on image signal processing methods.

[0065] It should be noted that data annotation of laryngeal images is a time-consuming and necessary procedure. In target detection tasks, it is often necessary to annotate image data frame by frame. However, the labeling module 2 provided in this embodiment of the invention does not require frame-by-frame annotation of image data when annotating. This embodiment of the invention proposes a low-workload annotation assistance technique. First, the conversion unit 21 uses Fourier transform to convert the image information of the laryngoscope image data into spectral information. Since the signal frequency is relatively high at the image boundary, the target can be initially identified based on this characteristic. Based on the continuity of the target position in each frame between images, and relying on the image spectral information obtained by the conversion unit 21, this embodiment of the invention uses the annotation assistance unit 22 to build a lightweight convolutional neural network and train a real-time annotation assistance component 220. The annotation assistance component 220 can automatically track the identified target in subsequent frames based on the number of already annotated image frames during annotation. The annotator only needs to correct the annotation of the specific frames where the annotation of the annotation assistance component 220 has deviated. Using this annotation aid component 220, annotators only need to use the selection unit 23 to select target locations for a small number of frames of image data, which greatly reduces the workload compared to the previous frame-by-frame annotation method.

[0066] Meanwhile, considering the complexity of real-world medical scenarios, this embodiment of the invention designs a label quality assessment method. Based on analyzing the contribution of each label to the model's correct predictions, an adaptive label quality assessment algorithm is designed. This adaptive algorithm completes the label quality assessment, and then, through weight design in the aforementioned neural network training, it makes greater use of high-quality labels and data, improving the model's robustness to different specific scenarios.

[0067] like Figure 1 As shown, the endotracheal intubation system also includes an image processing module 3, which is used to process the target tracheal image and simulate special clinical scenarios.

[0068] like Figure 1 As shown, the image processing module 3 includes: a processing unit 31, used to process the target tracheal image based on a data augmentation method; and a simulation unit 32, used to simulate specific clinical scenarios based on a generative adversarial network method. For example, specific clinical scenarios may include hemoptysis in the throat, foreign bodies or tumors in the pharynx, and anatomical deformities of the pharynx.

[0069] It should be noted that deep learning models tend to be trained using large amounts of data, which is problematic when actual medical data is scarce. This invention employs data augmentation methods to process target tracheal images. Specifically, by adjusting the input image to achieve different levels of contrast, brightness, and hue, and simulating enhancement effects for different situations, the usability of the data can be effectively improved. In addition to traditional image processing methods such as rotation, flipping, translation, and noise addition, this invention proposes a special image processing method to simulate data from clinically challenging scenarios, such as hemoptysis, foreign bodies or tumors, and anatomical deformities. These special scenarios are difficult to simulate using traditional image processing methods alone. To generate such scarce data, generative adversarial network algorithms, such as Generative Adversarial Networks (GANs), can be incorporated. These GANs rely on the competition between two neural networks to automatically simulate and generate scarce special scenario data. The generated image data is then used to expand the images of these clinically difficult-to-identify scenarios, such as hemoptysis, foreign bodies or tumors, and anatomical deformities. This helps improve the model's performance and robustness in these special scenarios.

[0070] like Figure 1 As shown, the endotracheal intubation system also includes a model building module 4, which is used to build an artificial intelligence model based on the processed target tracheal image, identify the trachea and esophagus based on the artificial intelligence model, and issue an alarm for incorrect intubation position.

[0071] like Figure 1As shown, the model building module 4 includes: a model training unit 41, which is used to input the target trachea image processed by the image processing module 3 into the artificial intelligence model, use the maximum likelihood function as the loss function to train the artificial intelligence model, and output the trained artificial intelligence model; and a model recognition unit 42, which is used to recognize the trachea and esophagus based on the trained artificial intelligence model and to alarm for incorrect intubation position.

[0072] It should be noted that, in order to identify the trachea and esophagus and to issue an alarm for incorrect intubation placement, this embodiment of the invention constructs an artificial intelligence model. Specifically, the target trachea image processed by processing unit 31 is input into the artificial intelligence model. Based on human anatomical organs and considering the relative positions of each organ, a maximum likelihood-likelihood-based evaluation method is proposed, and the maximum likelihood function is used as the loss function to train the artificial intelligence model. By multiplying the position probabilities of each target, this evaluation method, compared to single-target recognition that relies solely on image features, incorporates information about the relative positions between targets, resulting in higher accuracy in recognizing multiple targets with known relative positions. Furthermore, since the target positions are necessarily continuous in time, constraints are designed for neural network training based on this, and images from consecutive frames are used for recognition during training to increase recognition accuracy. For example, the artificial intelligence model can combine convolutional neural network architectures with time series network model architectures such as LSTM and Transformer. This study utilizes a convolutional neural network (CNN) architecture for image feature extraction, employs a time-series model to analyze the correlation between consecutive frames, and compares various neural network structures for target location identification in images. Model ensemble techniques are used to combine high-performance model architectures, achieving near-complex model performance through multiple lightweight models. For hyperparameter selection, a neural network hyperparameter grid search technique is employed to automatically optimize hyperparameters. The established AI model adopts a lightweight design, employing model compression techniques such as structure optimization, quantization, pruning, and distillation to reduce computation time, enabling real-time use in real-world clinical scenarios under significant time pressure. The established AI model can identify the location of the trachea and esophagus and issue alarms for incorrect intubation placement, thus guiding medical staff in intubation and providing warnings for incorrect placement.

[0073] Compared with the prior art, the endotracheal intubation system provided in this embodiment of the invention has the following beneficial effects:

[0074] 1. Identification markers for the trachea and esophagus.

[0075] The system uses image acquisition module 1 to acquire laryngoscope images, labeling module 2 to annotate the target trachea image, and image processing module 3 to process the target trachea image and simulate specific clinical scenarios. An artificial intelligence model is constructed using the processed target trachea image. Based on this model, the system autonomously labels the locations of the trachea and esophagus, guiding medical staff to perform corresponding intubation procedures via audio or video. Through real-time clinical annotation and guidance, clear voice and video guidance is provided to medical staff, assisting frontline medical personnel in intubation operations and reducing the risk of operational errors.

[0076] 2. Active cannulation position error detection and alarm.

[0077] By leveraging artificial intelligence models to analyze laryngoscopy images, the system proactively detects the condition of the larynx. When an error in tube placement is identified, it promptly alerts medical staff for correction. This proactive error detection eliminates the need for medical personnel to perform examinations, saving valuable time and energy in high-pressure, emergency clinical settings. It also reduces the probability of errors during operation, mitigating losses caused by medical malpractice.

[0078] 3. Existing patented technologies mostly involve direct hardware redesign of the laryngoscope itself. Such designs require large-scale replacement of medical equipment due to hardware modifications. However, the embodiments of this invention propose a cross-device solution based on existing medical equipment, which has a lower barrier to entry for practical application.

[0079] Figure 2 A flowchart of an endotracheal intubation method provided by an embodiment of the present invention is shown. Figure 3 This diagram illustrates the overall framework of an endotracheal intubation method provided by an embodiment of the present invention.

[0080] like Figure 2 and Figure 3 As shown, this embodiment of the invention also provides a method for endotracheal intubation, the method comprising:

[0081] Step 1: Acquire laryngoscope image data.

[0082] Step 2: Annotate the target trachea image based on the laryngoscope image data.

[0083] It should be noted that the target tracheal image labeled based on laryngoscopy image data includes:

[0084] Step 2.1: Convert the image information of the laryngoscope image data into spectral information based on Fourier transform;

[0085] Step 2.2: Since the signal frequency is relatively high at the image boundary, based on this, a pixel-level annotation auxiliary component is trained using a neural network according to the above spectrum information;

[0086] Step 2.3: Select the target tracheal image using the above annotation auxiliary components;

[0087] Step 2.4: Automatically segment the border details of the target trachea image using image signal processing methods.

[0088] Step 3: Process the target tracheal image and simulate a special clinical scenario.

[0089] It should be noted that the processing of the target tracheal image and the simulation of specific clinical scenarios include:

[0090] Step 3.1: Process the target tracheal image using data augmentation methods;

[0091] Step 3.2: Simulate specific clinical scenarios based on generative adversarial networks.

[0092] For example, specific clinical scenarios could include a foreign body or tumor in the pharynx, hemoptysis in the larynx, or anatomical deformity of the pharynx.

[0093] Step 4: Construct an artificial intelligence model based on the processed target trachea image, identify the trachea and esophagus based on the artificial intelligence model, and issue an alarm for incorrect intubation position.

[0094] It should be noted that an artificial intelligence model is built based on the processed target tracheal image. This model identifies the trachea and esophagus and issues an alarm for incorrect intubation placement, including:

[0095] Step 4.1: Input the processed target trachea image into the artificial intelligence model, use the maximum likelihood function as the loss function to train the artificial intelligence model, and output the trained artificial intelligence model.

[0096] Step 4.2: Identify the trachea and esophagus based on the trained artificial intelligence model, and issue an alarm for incorrect intubation position.

[0097] Compared with the prior art, the beneficial effects of the endotracheal intubation method provided by the present invention are the same as those of the endotracheal intubation system described in the above technical solution, and will not be repeated here.

[0098] In addition, this invention also provides an electronic device, including a bus, a transceiver, a memory, a processor, and a computer program stored in the memory and executable on the processor. The transceiver, the memory, and the processor are connected via the bus. When the computer program is executed by the processor, it implements the various processes of the above-described endotracheal intubation method embodiment and achieves the same technical effect. To avoid repetition, it will not be described again here.

[0099] For details, see Figure 4 As shown, this embodiment of the invention also provides an electronic device, which includes a bus 1110, a processor 1120, a transceiver 1130, a bus interface 1140, a memory 1150, and a user interface 1160.

[0100] In this embodiment of the invention, the electronic device further includes a computer program stored in a memory 1150 and executable on a processor 1120, wherein the computer program, when executed by the processor 1120, implements the various processes of the above-described embodiment of the endotracheal intubation method.

[0101] Transceiver 1130 is used to receive and send data under the control of processor 1120.

[0102] In this embodiment of the invention, a bus architecture (represented by bus 1110) is used. Bus 1110 may include any number of interconnected buses and bridges. Bus 1110 connects various circuits, including one or more processors represented by processor 1120 and memory represented by memory 1150.

[0103] Bus 1110 represents one or more of several types of bus architectures, including memory buses and memory controllers, peripheral buses, Accelerated Graphics Port (AGP), processors, or local buses using any bus architecture from various bus architectures. As an example and not a limitation, such architectures include: Industry Standard Architecture (ISA) buses, Micro Channel Architecture (MCA) buses, Enhanced ISA (EISA) buses, Video Electronics Standards Association (VESA) buses, and Peripheral Component Interconnect (PCI) buses.

[0104] The processor 1120 can be an integrated circuit chip with signal processing capabilities. In implementation, the steps of the above method embodiments can be completed by integrated logic circuits in the processor hardware or by instructions in software form. The processors mentioned above include: general-purpose processors, central processing units (CPUs), network processors (NPs), digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), complex programmable logic devices (CPLDs), programmable logic arrays (PLAs), microcontroller units (MCUs) or other programmable logic devices, discrete gates, transistor logic devices, and discrete hardware components. They can implement or execute the methods, steps, and logic block diagrams disclosed in the embodiments of this invention. For example, the processor can be a single-core processor or a multi-core processor, and the processor can be integrated on a single chip or located on multiple different chips.

[0105] Processor 1120 can be a microprocessor or any conventional processor. The method steps disclosed in the embodiments of the present invention can be directly executed by a hardware decoding processor, or executed by a combination of hardware and software modules in the decoding processor. The software modules can reside in readable storage media known in the art, such as Random Access Memory (RAM), Flash Memory, Read-Only Memory (ROM), Programmable Read-Only Memory (PROM), Erasable Programmable Read-Only Memory (EPROM), registers, etc. The readable storage medium is located in the memory, and the processor reads the information in the memory and, in conjunction with its hardware, completes the steps of the above method.

[0106] Bus 1110 can also connect various other circuits, such as peripheral devices, voltage regulators, or power management circuits. Bus interface 1140 provides an interface between bus 1110 and transceiver 1130, all of which are well known in the art. Therefore, the embodiments of the present invention will not be described further.

[0107] Transceiver 1130 can be a single element or multiple elements, such as multiple receivers and transmitters, providing a unit for communicating with various other devices over a transmission medium. For example, transceiver 1130 receives external data from other devices, and transceiver 1130 is used to send data processed by processor 1120 to other devices. Depending on the nature of the computer system, a user interface 1160 may also be provided, such as a touchscreen, physical keyboard, monitor, mouse, speaker, microphone, trackball, joystick, or stylus.

[0108] It should be understood that, in embodiments of the present invention, memory 1150 may further include memory remotely configured relative to processor 1120, and such remotely configured memory can be connected to a server via a network. One or more portions of the aforementioned network may be an ad hoc network, intranet, extranet, virtual private network (VPN), local area network (LAN), wireless local area network (WLAN), wide area network (WAN), wireless wide area network (WWAN), metropolitan area network (MAN), Internet, public switched telephone network (PSTN), ordinary old-style telephone service (POTS), cellular telephone network, wireless network, Wi-Fi network, and combinations of two or more of the aforementioned networks. For example, cellular telephone networks and wireless networks can be Global System for Mobile Communications (GSM), Code Division Multiple Access (CDMA), WiMAX, General Packet Radio Service (GPRS), Wideband Code Division Multiple Access (WCDMA), Long Term Evolution (LTE), LTE Frequency Division Duplex (FDD), LTE Time Division Duplex (TDD), Advanced Long Term Evolution (LTE-A), Universal Mobile Telecommunications System (UMTS), Enhanced Mobile Broadband (eMBB), Massive Machine Type Communication (mMTC), Ultra Reliable Low Latency Communications (uRLLC), etc.

[0109] It should be understood that the memory 1150 in the embodiments of the present invention may be volatile memory or non-volatile memory, or may include both volatile memory and non-volatile memory. Non-volatile memory includes: read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), or flash memory.

[0110] Volatile memory includes random access memory (RAM), which serves as an external cache. By way of example, but not limitation, many forms of RAM are available, such as static random access memory (SRAM), dynamic random access memory (DRAM), synchronous dynamic random access memory (SDRAM), double data rate synchronous dynamic random access memory (DDRSDRAM), enhanced synchronous dynamic random access memory (ESDRAM), synchronous linked dynamic random access memory (SLDRAM), and direct memory bus RAM (DRRAM). The memory 1150 of the electronic device described in this embodiment includes, but is not limited to, the above-described and any other suitable types of memory.

[0111] In this embodiment of the invention, the memory 1150 stores the following elements of the operating system 1151 and the application 1152: executable modules, data structures, or subsets thereof, or extended sets thereof.

[0112] Specifically, the operating system 1151 includes various system programs, such as a framework layer, a core library layer, and a driver layer, used to implement various basic business functions and handle hardware-based tasks. The application program 1152 includes various applications, such as a media player and a browser, used to implement various application functions. Programs implementing the methods of this embodiment of the invention can be included in the application program 1152. The application program 1152 includes applets, objects, components, logic, data structures, and other computer system executable instructions that perform specific tasks or implement specific abstract data types.

[0113] In addition, this embodiment of the invention also provides a computer-readable storage medium storing a computer program thereon. When the computer program is executed by a processor, it implements the various processes of the above-described endotracheal intubation method embodiment and achieves the same technical effect. To avoid repetition, it will not be described again here.

[0114] Computer-readable storage media include: permanent and non-permanent, removable and non-removable media, which are tangible devices capable of retaining and storing instructions for use by an instruction execution device. Computer-readable storage media include: electronic storage devices, magnetic storage devices, optical storage devices, electromagnetic storage devices, semiconductor storage devices, and any suitable combination thereof. Computer-readable storage media include: phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), non-volatile random access memory (NVRAM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, optical disc read-only memory (CD-ROM), digital versatile optical disc (DVD) or other optical storage, magnetic tape storage, magnetic disk storage or other magnetic storage devices, memory sticks, mechanical encoding devices (e.g., punched cards or raised structures in grooves on which instructions are recorded), or any other non-transfer medium that can be used to store information accessible by a computing device. As defined in the embodiments of the present invention, computer-readable storage media do not include temporary signals themselves, such as radio waves or other freely propagating electromagnetic waves, electromagnetic waves propagating through waveguides or other transmission media (e.g., light pulses passing through fiber optic cables), or electrical signals transmitted through wires.

[0115] In the several embodiments provided in this application, it should be understood that the disclosed apparatus, electronic devices, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative. For instance, the division of modules or units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. In addition, the mutual coupling or direct coupling or communication connection shown or discussed may be indirect coupling or communication connection through some interfaces, devices, or units, or it may be an electrical, mechanical, or other form of connection.

[0116] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to solve the problems addressed by the embodiments of the present invention, depending on actual needs.

[0117] Furthermore, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0118] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the embodiments of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (including: a personal computer, a server, a data center, or other network device) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media listed above that can store program code.

[0119] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. An endotracheal intubation system, characterized in that, include: Image acquisition module, used to acquire laryngoscope image data; The tagging module is used to annotate the target tracheal image based on the laryngoscope image data; The image processing module is used to process the target tracheal image and simulate special clinical scenarios, including: foreign bodies or tumors in the pharynx, hemoptysis in the larynx, and anatomical deformities of the pharynx. The model building module is used to build an artificial intelligence model based on the processed target tracheal image, identify the trachea and esophagus based on the artificial intelligence model, and issue an alarm for incorrect intubation position. The image processing module includes: The processing unit is used to process the target tracheal image based on a data augmentation method; The simulation unit is used to simulate the specific clinical scenario based on the generative adversarial network method.

2. The endotracheal intubation system according to claim 1, characterized in that, The tag module includes: The conversion unit is used to convert the image information of the laryngoscope image data into spectral information based on Fourier transform; A labeling assistance unit is used to train a pixel-level labeling assistance component based on a neural network according to the spectral information. The selection unit is used to select the target tracheal image based on the annotation auxiliary component; The border annotation unit is used to automatically segment the border details of the target trachea image based on image signal processing methods.

3. The endotracheal intubation system according to claim 1, characterized in that, The model building module includes: The model training unit is used to input the processed target tracheal image into the artificial intelligence model for training, and output the trained artificial intelligence model. The model recognition unit is used to identify the trachea and esophagus based on the trained artificial intelligence model and to issue an alarm for incorrect intubation position.

4. The endotracheal intubation system according to claim 3, characterized in that, The model training unit is used to input the processed target tracheal image into the artificial intelligence model for training, including: The model training unit is used to input the processed target tracheal image into the artificial intelligence model and train the artificial intelligence model using the maximum likelihood function as the loss function.

5. A method for endotracheal intubation, characterized in that, include: Step 1: Acquire laryngoscope image data; Step 2: Annotate the target trachea image based on the laryngoscope image data; Step 3: Process the target tracheal image and simulate special clinical scenarios, including: foreign bodies or tumors in the pharynx, hemoptysis in the larynx, and anatomical deformities of the pharynx; Step 4: Construct an artificial intelligence model based on the processed target trachea image, identify the trachea and esophagus based on the artificial intelligence model, and issue an alarm for incorrect intubation position; Step 3: Processing the target tracheal image and simulating specific clinical scenarios includes: Step 3.1: Process the target tracheal image using data augmentation methods; Step 3.2: Simulate the specific clinical scenario based on the generative adversarial network method.

6. The endotracheal intubation method according to claim 5, characterized in that, Step 2: Annotating the target tracheal image based on the laryngoscopy image data includes: Step 2.1: Convert the image information of the laryngoscope image data into spectral information based on Fourier transform; Step 2.2: Based on the spectral information, train a pixel-level annotation auxiliary component using a neural network; Step 2.3: Select the target tracheal image based on the annotation auxiliary component; Step 2.4: Automatically segment the border details of the target trachea image based on image signal processing methods.

7. A method for endotracheal intubation according to claim 5, characterized in that, Step 4: Construct an artificial intelligence model based on the processed target tracheal image, identify the trachea and esophagus based on the artificial intelligence model, and issue an alarm for incorrect intubation position, including: Step 4.1: Input the processed target tracheal image into the artificial intelligence model, train the artificial intelligence model using the maximum likelihood function as the loss function, and output the trained artificial intelligence model; Step 4.2: Identify the trachea and esophagus based on the trained artificial intelligence model, and issue an alarm for incorrect intubation position.

8. An electronic device comprising a bus, a transceiver, a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the transceiver, the memory, and the processor are connected via the bus, characterized in that, When the computer program is executed by the processor, it implements a tracheal intubation method as described in any one of claims 5 to 7.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the steps of a tracheal intubation method as described in any one of claims 5 to 7.

Images