Real-time auxiliary method and device for double-cavity tracheal intubation and electronic equipment

Abstract

The invention provides a real-time auxiliary method and device for double-cavity tracheal intubation and electronic equipment, and relates to the technical field of medical auxiliary decision making. The method comprises the steps that a tracheal intubation operation video stream is obtained in real time through a medical endoscope camera system; carrying out key frame extraction on the tracheal intubation operation video stream, and determining effective frames by adopting a dynamic threshold value method; and inputting the effective frame into a double-lumen tracheal intubation auxiliary identification model, outputting a double-lumen tracheal key anatomical structure in real time, and assisting in generating a double-lumen tracheal intubation strategy. According to the invention, individualized anesthesia scheme optimization can be carried out based on clinical data of a patient, key indexes such as a lumen position and ventilation parameters can be monitored in real time, and dynamic adjustment suggestions can be provided.

Description

Technical Field

[0001] The present invention relates to the technical field of medical decision-making assistance, and in particular to a real-time assistance method, device and electronic equipment for double-lumen endotracheal intubation. Background Art

[0002] Traditional double-lumen intubation decisions rely primarily on the anesthesiologist's personal experience and judgment. This experience-based decision-making model has several limitations. First, in complex cases such as those with advanced age, concurrent cardiopulmonary insufficiency, and airway anatomy, relying solely on experience can make it difficult to accurately assess intubation difficulty and develop the optimal plan. Second, in emergencies (such as a sharp drop in blood oxygen saturation or bronchospasm), experience-based decisions can result in delayed response. Third, differences in experience among physicians of different levels can lead to deviations in key decision points such as intubation positioning and depth assessment, increasing the risk of surgical delays and perioperative complications. Especially in emergency situations such as nighttime emergency surgery, physician fatigue can further affect the accuracy of judgments. These factors can jeopardize patient safety.

[0003] Therefore, a real-time auxiliary method, device and electronic equipment for double-lumen endotracheal intubation are proposed. Summary of the Invention

[0004] This specification provides a real-time auxiliary method, device and electronic equipment for double-lumen endotracheal intubation, which can not only optimize individualized anesthesia plans based on the patient's clinical data, but also monitor key indicators such as lumen position and ventilation parameters in real time and provide dynamic adjustment suggestions.

[0005] This specification provides a real-time assisted double-lumen endotracheal intubation method, including:

[0006] Real-time video stream of tracheal intubation surgery is obtained through a medical endoscope camera system;

[0007] Extracting key frames from the endotracheal intubation surgery video stream and determining valid frames using a dynamic threshold method;

[0008] The valid frames are input into a double-lumen tracheal intubation auxiliary recognition model, and the key anatomical structures of the double-lumen trachea are output in real time to assist in generating a double-lumen tracheal intubation strategy.

[0009] Optionally, the determining a valid frame by using a dynamic threshold method includes:

[0010] Analyzing the clarity and / or exposure level of each key frame, and / or analyzing the displacement between the key frames;

[0011] Determine whether the clarity and / or the exposure level and / or the displacement meet a preset value, and when the clarity and / or the exposure level and / or the displacement meet the preset value, determine it as a valid frame.

[0012] Optionally, before inputting the valid frame into the double-lumen tracheal intubation auxiliary recognition model and outputting the key anatomical structure of the double-lumen trachea in real time to assist in generating the double-lumen tracheal intubation strategy, the method includes: pixel standardization and color normalization of the valid frame, and medical enhancement processing.

[0013] Optionally, the construction of the double-lumen endotracheal tube auxiliary recognition model includes: using EfficientNet-B1 as the backbone network, whose composite scaling factor is width 1.0 / depth 1.1 / resolution 1.1, and subsequently cascading a position attention module and a channel attention module.

[0014] Optionally, the EfficientNet-B1 backbone network contains several levels of MBConv blocks:

[0015] Stage 1: Initial MBConv1 block, using a 3×3 kernel, 32 input channels, and 16 output channels;

[0016] Stage 2: MBConv4 block, using a 3×3 kernel, 16 input channels, and 24 output channels;

[0017] Stage 3: MBConv6 block, using a 3×3 kernel, 24 input channels, and 40 output channels;

[0018] Stage 4: MBConv6 block, using a 5×5 kernel, 40 input channels, and 80 output channels;

[0019] Stage 5: MBConv6 block, using a 5×5 kernel, 80 input channels, and 112 output channels;

[0020] Stage 6: MBConv6 block, using a 5×5 kernel, 112 input channels, and 192 output channels;

[0021] Stage 7: MBConv6 block using a 3×3 kernel, 192 input channels, and 320 output channels.

[0022] Optionally, the key anatomical structures of the double-lumen trachea include the oral cavity, epiglottis, glottis, main trachea, carina, right main trachea, and right secondary protuberance.

[0023] Optionally, the double-lumen endotracheal tube auxiliary recognition model is optimized using a FocalLoss loss function, specifically:

[0024] FL(p t )=-αt (1-p t ) γ log(p t )

[0025] Among them, the p t is the model's predicted probability of the key anatomical structure t of the double-lumen trachea, α t is a class balance factor used to balance the unbalanced effects of the key anatomical structures of the double-lumen trachea in the dataset, and γ is a modulation factor.

[0026] This specification provides a double-lumen endotracheal intubation real-time auxiliary device, including:

[0027] An acquisition module, used for acquiring a real-time video stream of an endotracheal intubation operation through a medical endoscope camera system;

[0028] An extraction module, configured to extract key frames from the endotracheal intubation surgery video stream and determine valid frames using a dynamic threshold method;

[0029] The auxiliary module is used to input the valid frame into the double-lumen tracheal intubation auxiliary recognition model, output the key anatomical structure of the double-lumen trachea in real time, and assist in generating the double-lumen tracheal intubation strategy.

[0030] Optionally, the extraction module includes:

[0031] Analyzing the clarity and / or exposure level of each key frame, and / or analyzing the displacement between the key frames;

[0032] Determine whether the clarity and / or the exposure level and / or the displacement meet a preset value, and when the clarity and / or the exposure level and / or the displacement meet the preset value, determine it as a valid frame.

[0033] Optionally, before the auxiliary module, it includes: pixel standardization and color normalization of the valid frame, and medical enhancement processing.

[0034] Optionally, the construction of the double-lumen endotracheal tube auxiliary recognition model includes: using EfficientNet-B1 as the backbone network, whose composite scaling factor is width 1.0 / depth 1.1 / resolution 1.1, and subsequently cascading a position attention module and a channel attention module.

[0035] Optionally, the EfficientNet-B1 backbone network contains several levels of MBConv blocks:

[0036] Stage 1: Initial MBConv1 block, using a 3×3 kernel, 32 input channels, and 16 output channels;

[0037] Stage 2: MBConv4 block, using a 3×3 kernel, 16 input channels, and 24 output channels;

[0038] Stage 3: MBConv6 block, using a 3×3 kernel, 24 input channels, and 40 output channels;

[0039] Stage 4: MBConv6 block, using a 5×5 kernel, 40 input channels, and 80 output channels;

[0040] Stage 5: MBConv6 block, using a 5×5 kernel, 80 input channels, and 112 output channels;

[0041] Stage 6: MBConv6 block, using a 5×5 kernel, 112 input channels, and 192 output channels;

[0042] Stage 7: MBConv6 block using a 3×3 kernel, 192 input channels, and 320 output channels.

[0043] Optionally, the key anatomical structures of the double-lumen trachea include the oral cavity, epiglottis, glottis, main trachea, carina, right main trachea, and right secondary protuberance.

[0044] Optionally, the double-lumen endotracheal tube auxiliary recognition model is optimized using a FocalLoss loss function, specifically:

[0045] FL(p t )=-α t (1-p t ) γ log(p t )

[0046] Among them, the p t is the model's predicted probability of the key anatomical structure t of the double-lumen trachea, α t is a class balance factor used to balance the unbalanced effects of the key anatomical structures of the double-lumen trachea in the dataset, and γ is a modulation factor.

[0047] This specification also provides an electronic device, wherein the electronic device includes:

[0048] processor; and,

[0049] A memory storing computer executable instructions, which, when executed, cause the processor to perform any of the above methods.

[0050] This specification also provides a computer-readable storage medium, wherein the computer-readable storage medium stores one or more programs, and when the one or more programs are executed by a processor, any of the above methods is implemented.

[0051] In the present invention, the real-time recognition system of airway anatomical structure based on deep learning can accurately identify seven key anatomical structures including the oral cavity, epiglottis, glottis, main trachea, carina, right main trachea, and right secondary protuberance by performing real-time analysis of the intubation process video, providing real-time guidance and auxiliary judgment for clinicians. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0053] Figure 1 A schematic diagram of the principle of a real-time auxiliary method for double-lumen endotracheal intubation provided in an embodiment of this specification;

[0054] Figure 2 A schematic structural diagram of a double-lumen endotracheal intubation real-time assistance device provided in an embodiment of this specification;

[0055] Figure 3 A schematic diagram of the structure of an electronic device provided in an embodiment of this specification;

[0056] Figure 4 A schematic diagram of a computer-readable medium provided in accordance with an embodiment of this specification. DETAILED DESCRIPTION

[0057] The following description is intended to disclose the present invention so that those skilled in the art can implement the present invention. The preferred embodiments described below are for illustrative purposes only, and those skilled in the art will readily appreciate other obvious variations. The basic principles of the present invention defined in the following description may be applied to other embodiments, variations, improvements, equivalents, and other technical solutions that do not depart from the spirit and scope of the present invention.

[0058] The following is combined with Figure 1-4 The exemplary embodiments of the present invention are described more fully. However, the exemplary embodiments can be implemented in various forms, and the present invention should not be construed as being limited to the embodiments set forth herein. On the contrary, providing these exemplary embodiments enables the present invention to be more comprehensive and complete, and more conveniently conveys the inventive concept to those skilled in the art. In the figures, the same reference numerals represent the same or similar elements, components, or parts, and thus their repeated description will be omitted.

[0059] Under the premise of being consistent with the technical concept of the present invention, the features, structures, characteristics or other details described in a specific embodiment do not exclude that they can be combined in one or more other embodiments in a suitable manner.

[0060] In the description of specific embodiments, the features, structures, characteristics, or other details of the present invention are described to enable those skilled in the art to fully understand the embodiments. However, this does not preclude those skilled in the art from practicing the technical solutions of the present invention without one or more of the specific features, structures, characteristics, or other details.

[0061] The flowcharts shown in the accompanying drawings are for illustrative purposes only and do not necessarily include all contents and operations / steps, nor must they be executed in the order described. For example, some operations / steps may be decomposed, while others may be combined or partially combined. Therefore, the actual execution order may vary depending on the actual situation.

[0062] The block diagrams shown in the accompanying drawings are merely functional entities and do not necessarily correspond to physically separate entities. That is, these functional entities may be implemented in software, in one or more hardware modules or integrated circuits, or in different networks and / or processor devices and / or microcontroller devices.

[0063] The term "and / or" or "and / or" includes all combinations of any one or more of the associated listed items.

[0064] Figure 1 A schematic diagram of the principle of a real-time auxiliary method for double-lumen endotracheal intubation provided in an embodiment of this specification, which may include:

[0065] S110: Real-time acquisition of tracheal intubation surgery video stream via a medical endoscope camera system;

[0066] In the specific embodiments of this specification, endotracheal intubation surgical videos were captured in a standard operating room environment. The acquisition process strictly adhered to medical standards and ethical requirements, and all data was desensitized. The video capture equipment used a medical-grade endoscopic camera system with a capture resolution of no less than 720p and a frame rate of at least 25fps to ensure image quality that meets medical diagnostic requirements. Standard operating room lighting was used to avoid image quality issues caused by insufficient or excessive light.

[0067] S120: extracting key frames from the tracheal intubation surgery video stream and determining valid frames using a dynamic threshold method;

[0068] Optionally, the S120 includes:

[0069] Analyzing the clarity and / or exposure level of each key frame, and / or analyzing the displacement between the key frames;

[0070] Determine whether the clarity and / or the exposure level and / or the displacement meet a preset value, and when the clarity and / or the exposure level and / or the displacement meet the preset value, determine it as a valid frame.

[0071] In a specific embodiment of the present specification, qualified frames with Laplace variance ≥ 0.85 and grayscale histogram exposure values ​​in the range of [0.3, 0.7] are calculated, and blurred frames with inter-frame displacement greater than 5 pixels are detected using the optical flow method.

[0072] Optionally, before S130, the step includes: standardizing pixels and colors of the valid frame, and performing medical enhancement processing.

[0073] In this specific implementation, the active frame is resized to the default size of 224×224 pixels, which is the default size for this model in ImageNet. The active frame is processed through a standardized color normalization procedure with a mean of [0.485, 0.456, 0.406] and a standard deviation of [0.229, 0.224, 0.225] to reduce the common lighting and equipment variations in medical image acquisition.

[0074] Data augmentation includes:

[0075] 1. Limited random rotation (±15°) to simulate different shooting angles;

[0076] 2. Slight brightness and contrast adjustments (brightness factor range is 0.9-1.1, contrast factor range is 0.9-1.1);

[0077] 3. Moderate random cropping and scaling (scaling factor range 0.95-1.05);

[0078] 4.50% probability of horizontal flip.

[0079] These augmentation techniques are carefully calibrated to improve model generalization while avoiding introducing artifacts that could compromise diagnostic features.

[0080] Optionally, the key anatomical structures of the double-lumen trachea include the oral cavity, epiglottis, glottis, main trachea, carina, right main trachea, and right secondary protuberance.

[0081] In the specific implementation of this specification, the oral category mainly includes the anatomical structures of the entrance area such as the oral cavity and tongue; the epiglottis category focuses on the morphological characteristics of the epiglottis cartilage; the glottis category needs to clearly show the opening status of the vocal cords and glottis; the main trachea category focuses on the lumen characteristics of the tracheal trunk; the carina category focuses on the bifurcation morphology of the tracheal carina; the right main trachea category focuses on the characteristics of the right main bronchus; and the right secondary protuberance category needs to identify the characteristics of the right secondary bronchial branches.

[0082] S130: Input the valid frame into a double-lumen tracheal intubation auxiliary recognition model, output the key anatomical structure of the double-lumen trachea in real time, and assist in generating a double-lumen tracheal intubation strategy.

[0083] Optionally, the construction of the double-lumen endotracheal tube auxiliary recognition model includes: using EfficientNet-B1 as the backbone network, whose composite scaling factor is width 1.0 / depth 1.1 / resolution 1.1, and subsequently cascading a position attention module and a channel attention module.

[0084] In the specific implementation of this specification, EfficientNet-B1 is selected as the basic network architecture. The network optimizes the network width, depth and resolution through a composite scaling method, achieving high recognition accuracy while maintaining a small model size.

[0085] Optionally, the EfficientNet-B1 backbone network contains several levels of MBConv blocks:

[0086] Stage 1: Initial MBConv1 block, using a 3×3 kernel, 32 input channels, and 16 output channels;

[0087] Stage 2: MBConv4 block, using a 3×3 kernel, 16 input channels, and 24 output channels;

[0088] Stage 3: MBConv6 block, using a 3×3 kernel, 24 input channels, and 40 output channels;

[0089] Stage 4: MBConv6 block, using a 5×5 kernel, 40 input channels, and 80 output channels;

[0090] Stage 5: MBConv6 block, using a 5×5 kernel, 80 input channels, and 112 output channels;

[0091] Stage 6: MBConv6 block, using a 5×5 kernel, 112 input channels, and 192 output channels;

[0092] Stage 7: MBConv6 block using a 3×3 kernel, 192 input channels, and 320 output channels.

[0093] In the specific implementation of this specification, the network body is composed of 7 MBConv blocks, and each module integrates the dual optimization mechanism of depth-wise separable convolution and squeeze excitation (SE): depth-wise separable convolution uses 3×3 or 5×5 kernels to reduce computational complexity by separating spatial and channel convolution operations; the SE module compresses features through global average pooling, builds a bottleneck structure through the fully connected layer, and generates a channel weight matrix to achieve dynamic calibration of feature channels.

[0094] The position attention module (PAM) constructs the query matrix, key matrix and value matrix, calculates the spatial attention map to enhance the spatial correlation features, and combines the residual connection to retain the original feature details; the channel attention module (CAM) generates the channel relationship matrix by reshaping the feature dimension, highlighting the diagnosis-related channel features.

[0095] The model uses 1×1 convolutions for spatially preserving dimensionality reduction, followed by batch normalization and ReLU activation. After global average pooling, the feature vector is obtained. A two-level fully connected layer with a decreasing dropout rate is used to output a probability distribution for seven anatomical structures. This structure effectively controls model complexity while maintaining classification accuracy.

[0096] Optionally, the double-lumen endotracheal tube auxiliary recognition model is optimized using a FocalLoss loss function, specifically:

[0097] FL(p t )=-α t (1-p t ) γ log(p t )

[0098] Among them, the p t is the model's predicted probability of the key anatomical structure t of the double-lumen trachea, α t is a class balance factor used to balance the unbalanced effects of the key anatomical structures of the double-lumen trachea in the dataset, and γ is a modulation factor.

[0099] In the specific implementation of this specification, when γ = 0, Focal Loss degenerates into weighted cross entropy loss; when γ increases, the loss contribution to easy-to-classify samples (high confidence predictions) will decrease, and the model will pay more attention to difficult-to-classify samples.

[0100] The present invention is a real-time airway anatomical structure recognition system based on deep learning. By performing real-time analysis on the intubation process video, it can accurately identify seven key anatomical structures including the oral cavity, epiglottis, glottis, main trachea, carina, right main trachea, and right secondary protuberance, providing real-time guidance and auxiliary judgment for clinicians.

[0101] Figure 2 A schematic diagram of the principle of a double-lumen endotracheal intubation real-time assistance device provided in an embodiment of this specification, the device may include:

[0102] An acquisition module 10 is used to acquire a video stream of a tracheal intubation operation in real time through a medical endoscope camera system;

[0103] An extraction module 20 is used to extract key frames from the endotracheal intubation surgery video stream and determine valid frames using a dynamic threshold method;

[0104] The auxiliary module 30 is used to input the valid frame into the double-lumen tracheal intubation auxiliary recognition model, output the key anatomical structure of the double-lumen trachea in real time, and assist in generating the double-lumen tracheal intubation strategy.

[0105] Optionally, the extraction module 20 includes:

[0106] Analyzing the clarity and / or exposure level of each key frame, and / or analyzing the displacement between the key frames;

[0107] Determine whether the clarity and / or the exposure level and / or the displacement meet a preset value, and when the clarity and / or the exposure level and / or the displacement meet the preset value, determine it as a valid frame.

[0108] Optionally, before the auxiliary module 30 , the following steps are included: pixel standardization and color normalization of the valid frame, and medical enhancement processing.

[0109] Optionally, the construction of the double-lumen endotracheal tube auxiliary recognition model includes: using EfficientNet-B1 as the backbone network, whose composite scaling factor is width 1.0 / depth 1.1 / resolution 1.1, and subsequently cascading a position attention module and a channel attention module.

[0110] Optionally, the EfficientNet-B1 backbone network contains several levels of MBConv blocks:

[0111] Stage 1: Initial MBConv1 block, using a 3×3 kernel, 32 input channels, and 16 output channels;

[0112] Stage 2: MBConv4 block, using a 3×3 kernel, 16 input channels, and 24 output channels;

[0113] Stage 3: MBConv6 block, using a 3×3 kernel, 24 input channels, and 40 output channels;

[0114] Stage 4: MBConv6 block, using a 5×5 kernel, 40 input channels, and 80 output channels;

[0115] Stage 5: MBConv6 block, using a 5×5 kernel, 80 input channels, and 112 output channels;

[0116] Stage 6: MBConv6 block, using a 5×5 kernel, 112 input channels, and 192 output channels;

[0117] Stage 7: MBConv6 block using a 3×3 kernel, 192 input channels, and 320 output channels.

[0118] Optionally, the key anatomical structures of the double-lumen trachea include the oral cavity, epiglottis, glottis, main trachea, carina, right main trachea, and right secondary protuberance.

[0119] Optionally, the double-lumen endotracheal tube auxiliary recognition model is optimized using a FocalLoss loss function, specifically:

[0120] FL(p t )=-α t (1-p t ) γ log(p t )

[0121] Among them, the p t is the model's predicted probability of the key anatomical structure t of the double-lumen trachea, α t is a class balance factor used to balance the unbalanced effects of the key anatomical structures of the double-lumen trachea in the dataset, and γ is a modulation factor.

[0122] The functions of the device in the embodiment of the present invention have been described in the above method embodiment. Therefore, for details not fully described in this embodiment, please refer to the relevant description in the above embodiment and will not be repeated here.

[0123] Based on the same inventive concept, an embodiment of this specification also provides an electronic device.

[0124] The following describes an electronic device embodiment of the present invention, which can be considered a specific physical implementation of the method and apparatus embodiments of the present invention described above. Details described in the electronic device embodiment of the present invention should be considered supplementary to the above-mentioned method or apparatus embodiments; details not disclosed in the electronic device embodiment of the present invention can be implemented with reference to the above-mentioned method or apparatus embodiments.

[0125] Figure 3 This is a schematic diagram of the structure of an electronic device provided in the embodiment of this specification. Figure 3 The electronic device 300 according to this embodiment of the present invention will be described. Figure 3 The electronic device 300 shown is merely an example and should not limit the functions and scope of use of the embodiments of the present invention.

[0126] like Figure 3 As shown, electronic device 300 is implemented as a general-purpose computing device. Components of electronic device 300 may include, but are not limited to, at least one processing unit 310, at least one storage unit 320, a bus 330 connecting various system components (including storage unit 320 and processing unit 310), a display unit 340, and the like.

[0127] The storage unit stores program codes that can be executed by the processing unit 310, so that the processing unit 310 performs the steps according to various exemplary embodiments of the present invention described in the above processing method section of this specification. For example, the processing unit 310 can perform the following steps: Figure 1 Steps shown.

[0128] The storage unit 320 may include a readable medium in the form of a volatile storage unit, such as a random access memory unit (RAM) 3201 and / or a cache memory unit 3202 , and may further include a read-only memory unit (ROM) 3203 .

[0129] The storage unit 320 may also include a program / utility 3204 having a set (at least one) of program modules 3205, such program modules 3205 including but not limited to: an operating system, one or more application programs, other program modules and program data, each of which or some combination may include an implementation of a network environment.

[0130] Bus 330 may represent one or more of several types of bus structures, including a memory bus or memory controller, a peripheral bus, an accelerated graphics port, a processing unit, or a local bus using any of a variety of bus architectures.

[0131] The electronic device 300 may also communicate with one or more external devices 400 (e.g., keyboards, pointing devices, Bluetooth devices, etc.), one or more devices that enable viewers to interact with the electronic device 300, and / or any device that enables the electronic device 300 to communicate with one or more other computing devices (e.g., routers, modems, etc.). Such communication may be performed through an input / output (I / O) interface 350. Furthermore, the electronic device 300 may also communicate with one or more networks (e.g., a local area network (LAN), a wide area network (WAN), and / or a public network, such as the Internet) through a network adapter 360. The network adapter 360 may communicate with other modules of the electronic device 300 through the bus 330. It should be understood that although Figure 3 Not shown, other hardware and / or software modules may be used in conjunction with the electronic device 300, including but not limited to microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage systems.

[0132] Through the description of the above embodiments, it is easy for those skilled in the art to understand that the exemplary embodiments described in the present invention can be implemented by software, or by combining software with necessary hardware. Therefore, the technical solution according to the embodiment of the present invention can be embodied in the form of a software product, which can be stored in a computer-readable storage medium (which can be a CD-ROM, a USB flash drive, a mobile hard disk, etc.) or on a network, and includes a number of instructions to enable a computing device (which can be a personal computer, a server, or a network device, etc.) to execute the above method according to the present invention. When the computer program is executed by a data processing device, the computer-readable medium is enabled to implement the above method of the present invention, that is: Figure 1 The method shown.

[0133] Figure 4 A schematic diagram of a computer-readable medium provided in accordance with an embodiment of this specification.

[0134] accomplish Figure 1 The computer program of the method shown can be stored on one or more computer-readable media. The computer-readable medium can be a readable signal medium or a readable storage medium. The readable storage medium can be, for example, but not limited to, a system, device or component of electricity, magnetism, light, electromagnetic, infrared, or semiconductor, or any combination thereof. More specific examples (non-exhaustive list) of readable storage media include: an electrical connection with one or more wires, a portable disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination thereof.

[0135] The computer-readable storage medium may include a data signal propagated in baseband or as part of a carrier wave, wherein the readable program code is carried. The data signal propagated may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. The readable storage medium may also be any readable medium other than a readable storage medium, which may send, propagate, or transmit a program for use by or in conjunction with an instruction execution system, device, or component. The program code contained on the readable storage medium may be transmitted using any suitable medium, including but not limited to wireless, wired, optical cable, RF, etc., or any suitable combination thereof.

[0136] The program code for performing the operations of the present invention may be written in any combination of one or more programming languages, including object-oriented programming languages ​​such as Java, C++, and the like, and conventional procedural programming languages ​​such as "C" or similar programming languages. The program code may be executed entirely on the viewer computing device, partially on the viewer device, as a stand-alone software package, partially on the viewer computing device and partially on a remote computing device, or entirely on a remote computing device or server. In cases involving a remote computing device, the remote computing device may be connected to the viewer computing device through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computing device (e.g., through the Internet using an Internet service provider).

[0137] In summary, the present invention can be implemented in hardware, or in a software module running on one or more processors, or in a combination thereof. It will be appreciated by those skilled in the art that general data processing equipment such as a microprocessor or a digital signal processor (DSP) can be used in practice to implement some or all of the functions of some or all of the components in the embodiments of the present invention. The present invention can also be implemented as a device or apparatus program (e.g., a computer program and a computer program product) for executing a part or all of the methods described herein. Such a program for implementing the present invention can be stored on a computer-readable medium, or can have the form of one or more signals. Such a signal can be downloaded from an Internet website, or provided on a carrier signal, or provided in any other form.

[0138] The specific embodiments described above further illustrate the objectives, technical solutions, and beneficial effects of the present invention. It should be understood that the present invention is not inherently related to any specific computer, virtual device, or electronic device, and various general-purpose devices can also implement the present invention. The above description is only a specific embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention shall be included in the scope of protection of the present invention.

[0139] The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on the differences from other embodiments.

[0140] The foregoing is merely an embodiment of the present application and is not intended to limit the present application. For those skilled in the art, the present application may have various changes and variations. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present application should all be included within the scope of the claims of the present application.

Claims

1. A real-time auxiliary method for double-lumen endotracheal intubation, characterized in that: include: Real-time video stream of tracheal intubation surgery is obtained through a medical endoscope camera system; Extracting key frames from the endotracheal intubation surgery video stream and determining valid frames using a dynamic threshold method; The valid frames are input into a double-lumen tracheal intubation auxiliary recognition model, and the key anatomical structures of the double-lumen trachea are output in real time to assist in generating a double-lumen tracheal intubation strategy.

2. The real-time assistance method for double-lumen endotracheal intubation according to claim 1, characterized in that: The method of determining a valid frame by using a dynamic threshold method includes: Analyzing the clarity and / or exposure level of each key frame, and / or analyzing the displacement between the key frames; Determine whether the clarity and / or the exposure level and / or the displacement meet a preset value, and when the clarity and / or the exposure level and / or the displacement meet the preset value, determine it as a valid frame.

3. The real-time assistance method for double-lumen endotracheal intubation according to claim 2, characterized in that: The method includes inputting the valid frame into the double-lumen tracheal intubation auxiliary recognition model, outputting the key anatomical structure of the double-lumen trachea in real time, and assisting in generating the double-lumen tracheal intubation strategy, including: pixel standardization and color normalization of the valid frame, and medical enhancement processing.

4. The real-time assistance method for double-lumen endotracheal intubation according to claim 3, characterized in that: The construction of the double-lumen endotracheal tube auxiliary recognition model includes: using EfficientNet-B1 as the backbone network, with a composite scaling factor of width 1.0 / depth 1.1 / resolution 1.1, and subsequently cascading position attention modules and channel attention modules.

5. The real-time assistance method for double-lumen endotracheal intubation according to claim 4, characterized in that: The EfficientNet-B1 backbone network contains several levels of MBConv blocks: Stage 1: Initial MBConv1 block, using a 3×3 kernel, 32 input channels, and 16 output channels; Stage 2: MBConv4 block, using a 3×3 kernel, 16 input channels, and 24 output channels; Stage 3: MBConv6 block, using a 3×3 kernel, 24 input channels, and 40 output channels; Stage 4: MBConv6 block, using a 5×5 kernel, 40 input channels, and 80 output channels; Stage 5: MBConv6 block, using a 5×5 kernel, 80 input channels, and 112 output channels; Stage 6: MBConv6 block, using a 5×5 kernel, 112 input channels, and 192 output channels; Stage 7: MBConv6 block using a 3×3 kernel, 192 input channels, and 320 output channels.

6. The real-time assistance method for double-lumen endotracheal intubation according to claim 5, characterized in that: The key anatomical structures of the double-lumen trachea include the oral cavity, epiglottis, glottis, main trachea, carina, right main trachea, and right secondary protuberance.

7. The real-time assistance method for double-lumen endotracheal intubation according to claim 6, characterized in that: The Focal Loss function is used to optimize the double-lumen endotracheal tube auxiliary recognition model. Specifically: FL(p t )=-a t (1-p t ) γ log(p t ) Among them, the p t is the model's predicted probability of the key anatomical structure t of the double-lumen trachea, α t is a class balance factor used to balance the unbalanced effects of the key anatomical structures of the double-lumen trachea in the dataset, and γ is a modulation factor.

8. A real-time assistance method for double-lumen endotracheal intubation, characterized in that: include: An acquisition module, used for acquiring a real-time video stream of an endotracheal intubation operation through a medical endoscope camera system; An extraction module, configured to extract key frames from the endotracheal intubation surgery video stream and determine valid frames using a dynamic threshold method; The auxiliary module is used to input the valid frame into the double-lumen tracheal intubation auxiliary recognition model, output the key anatomical structure of the double-lumen trachea in real time, and assist in generating the double-lumen tracheal intubation strategy.

9. An electronic device, wherein: The electronic device includes: processor; and, A memory storing computer executable instructions which, when executed, cause the processor to perform the method according to any one of claims 1 to 7.

10. A computer-readable storage medium, wherein: The computer-readable storage medium stores one or more programs, and when the one or more programs are executed by a processor, the method of any one of claims 1 to 7 is implemented.

Images