A Difficult Airway Assessment Method Based on a Large Model of Audio and Video Features

Abstract

The present invention discloses a difficult airway assessment method based on an audio and video feature large model, including calculating an adjustment coefficient of the fusion value of the difficult airway audio-visual feature data spatial mixed sequence and the difficult airway audio-visual feature data temporal mixed sequence through the fusion value of the difficult airway audio-visual feature data spatial mixed sequence and the difficult airway audio-visual feature data temporal mixed sequence. Adjusting the data sequence can achieve abnormal fluctuations in the fusion value of the difficult airway audio-visual feature data spatial mixed sequence and the difficult airway audio-visual feature data temporal mixed sequence, reduce the influence on the training data sequence of the large model for the fusion of the difficult airway audio-visual feature data spatial mixed sequence and the difficult airway audio-visual feature data temporal mixed sequence, and realize the assessment of the difficult airway with multiple parameters; and comprehensively considering the audio feature data sequence of the difficult airway and the video feature data sequence of the difficult airway, the dimensional calculation of multiple parameter spaces is realized.

Description

Technical Field

[0001] The present invention relates to the technical field of airway assessment, and particularly to a difficult airway assessment method based on an audio and video feature large model. Background Art

[0002] The assessment of difficult airways plays an important role in the analysis of tracheal intubability and tracheal intubation safety. If the assessment of difficult airways is inaccurate, there are potential safety hazards. Existing assessment methods for difficult airways include those based on subjective experience assessment, those based on audio analysis assessment, and those based on video analysis assessment, but their assessment dimensions are relatively single and do not fully utilize the multi-dimensionality of data.

[0003] With the development of artificial intelligence technology, large models have shown powerful learning capabilities. Therefore, the present invention proposes a difficult airway assessment method based on an audio and video feature large model, which fully utilizes the learning capabilities of the large model and the multi-dimensionality of audio and video data. Summary of the Invention

[0004] The purpose of this part is to outline some aspects of the embodiments of the present invention and briefly introduce some preferred embodiments. Simplifications or omissions may be made in this part, as well as in the abstract and title of the present application, to avoid obscuring the purpose of this part, the abstract, and the title, and such simplifications or omissions shall not be used to limit the scope of the present invention.

[0005] In view of the above existing problems, the present invention is proposed. Therefore, the present invention provides a difficult airway assessment method based on an audio and video feature large model to solve the problems raised in the background art.

[0006] To solve the above technical problems, the present invention provides the following technical solution: A difficult airway assessment method based on an audio and video feature large model, comprising:

[0007] Real-time collection of difficult airway state features through an airway audio collection system and an airway video collection system, and transmission and processing of the collected difficult airway state features by using a high-speed and low-latency data network communication method;

[0008] Constructing a difficult airway large model through a large model server, receiving the difficult airway state features transmitted by the high-speed and low-latency data network communication method, and performing current moment large model training on the difficult airway state features based on the difficult airway large model to obtain a predictive assessment of the difficult airway large model;

[0009] Based on the predictive assessment, establishing an auxiliary decision for the difficult airway large model and performing iterative optimization on the difficult airway large model through a real-time backpropagation mechanism.

[0010] As a preferred solution of the difficult airway assessment method based on the large model of audio and video features according to the present invention, wherein: the difficult airway state features are collected in real time through the airway audio acquisition system and the airway video acquisition system, and the collected difficult airway state features are transmitted and processed by means of high-speed and low-latency data network communication, including:

[0011] The processing method is to perform normalization processing and out-of-threshold data elimination processing on the data, and perform fusion large model training data sequence processing on the spatial mixed sequence of difficult airway audio and video feature data and the temporal mixed sequence of difficult airway audio and video feature data;

[0012] The difficult airway state features at least include the audio feature data sequence of the difficult airway and the video feature data sequence of the difficult airway.

[0013] As a preferred solution of the difficult airway assessment method based on the large model of audio and video features according to the present invention, wherein: the fusion large model training data sequence processing of the spatial mixed sequence of difficult airway audio and video feature data and the temporal mixed sequence of difficult airway audio and video feature data includes:

[0014] Calculate the spatial mixed sequence of the spatial mixed sequence of difficult airway audio and video feature data;

[0015] Calculate the fusion value of the temporal mixed sequence of difficult airway audio and video feature data, the spatial mixed sequence of difficult airway audio and video feature data, and the temporal mixed sequence of difficult airway audio and video feature data.

[0016] As a preferred solution of the difficult airway assessment method based on the large model of audio and video features according to the present invention, wherein: it further includes:

[0017] Calculate the fusion value adjustment coefficient of the spatial mixed sequence of difficult airway audio and video feature data and the temporal mixed sequence of difficult airway audio and video feature data;

[0018] Calculate the fusion large model training data sequence of the spatial mixed sequence of difficult airway audio and video feature data and the temporal mixed sequence of difficult airway audio and video feature data.

[0019] As a preferred solution of the difficult airway assessment method based on the large model of audio and video features according to the present invention, wherein: a difficult airway large model is constructed through the large model server, and the difficult airway state features transmitted by means of high-speed and low-latency data network communication are received, including:

[0020] The large model server constructs the difficult airway large model based on the difficult airway state features collected in the past and the difficult airway state features collected currently;

[0021] Among them, the difficult airway large model is privately and locally deployed using an open-source large model framework.

[0022] As a preferred solution of the difficult airway assessment method based on the large model of audio and video features according to the present invention, wherein: the current moment large model training is performed on the difficult airway state features based on the difficult airway large model to obtain the predictive evaluation of the difficult airway large model, including:

[0023] The difficult airway large model compares the actually collected difficult airway state features with the prediction evaluation results of the difficult airway large model, and iteratively improves the training accuracy of the difficult airway large model.

[0024] As a preferred solution of the difficult airway assessment method based on the large model of audio and video features according to the present invention, wherein: it further includes:

[0025] The difficult airway large model uses a local large model neural network to calculate the change trend of the difficult airway data sequence features, and evaluates the tracheal intubability and tracheal intubation safety of the actual difficult airway according to preset rules.

[0026] As a preferred solution of the difficult airway assessment method based on the large model of audio and video features according to the present invention, wherein: according to the prediction evaluation, an auxiliary decision of the difficult airway large model is established, including:

[0027] Using adaptive particle swarm optimization decision-making as the auxiliary decision of the difficult airway large model;

[0028] The adaptive particle swarm optimization decision-making obtains the operation safety score of the difficult airway according to the tracheal intubability and tracheal intubation safety of the actual difficult airway and the prediction evaluation results of the difficult airway large model.

[0029] Compared with the prior art, the beneficial effects of the invention are:

[0030] 1. By introducing the fusion value adjustment coefficient of the difficult airway audio-video feature data space hybrid sequence and the difficult airway audio-video feature data time hybrid sequence, the present invention calculates the fusion value adjustment coefficient of the difficult airway audio-video feature data space hybrid sequence and the difficult airway audio-video feature data time hybrid sequence through the fusion value of the difficult airway audio-video feature data space hybrid sequence and the difficult airway audio-video feature data time hybrid sequence. By adjusting the data sequence, abnormal fluctuations in the fusion value of the difficult airway audio-video feature data space hybrid sequence and the difficult airway audio-video feature data time hybrid sequence can be achieved, reducing the impact on the fusion large model training data sequence of the difficult airway audio-video feature data space hybrid sequence and the difficult airway audio-video feature data time hybrid sequence;

[0031] 2. The present invention comprehensively considers the audio feature data sequence and video feature data sequence of a difficult airway, and can realize the evaluation of a difficult airway with multiple parameters. At the same time, it calculates the spatial mixed sequence of the audio and video feature data of the difficult airway at the current sampling moment, and calculates the temporal mixed sequence of the audio and video feature data of the difficult airway at the current sampling moment, so as to realize the calculation of the fusion value of the spatial actual dimension of multiple parameters for the spatial mixed sequence and temporal mixed sequence of the audio and video feature data of the difficult airway, enabling the fusion large model of the spatial mixed sequence and temporal mixed sequence of the audio and video feature data of the difficult airway to fully consider the factors of time and space scale during the training process. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the following drawings are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings. Among them:

[0033] Figure 1 It is the overall flowchart of the method for evaluating a difficult airway based on an audio and video feature large model according to an embodiment of the present invention;

[0034] Figure 2 It is the flowchart for processing the training data sequence of the fusion large model of the spatial mixed sequence and temporal mixed sequence of the audio and video feature data of the difficult airway in the method for evaluating a difficult airway based on an audio and video feature large model according to an embodiment of the present invention;

[0035] Figure 3 It is the comparison diagram before and after the fusion of audio and video data in the method for evaluating a difficult airway based on an audio and video feature large model according to an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0036] In order to make the above objects, features, and advantages of the present invention more obvious and understandable, the following will provide a detailed description of the specific embodiments of the present invention with reference to the drawings of the specification. Obviously, the described embodiments are some embodiments of the present invention, rather than all embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0037] In the following description, numerous specific details are set forth in order to provide a thorough understanding of the present invention. However, the present invention may be practiced in other ways than those specifically described herein. Those skilled in the art can make similar extensions without departing from the spirit of the present invention. Therefore, the present invention is not limited by the specific embodiments disclosed below.

[0038] Secondly, the so-called "one embodiment" or "embodiment" herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation manner of the present invention. The appearances of "in one embodiment" in different places in this specification do not all refer to the same embodiment, nor are they separate or alternative embodiments that exclude each other.

[0039] The present invention is described in detail in conjunction with schematic diagrams. When detailing the embodiments of the present invention, for ease of explanation, the cross-sectional views showing the device structure will be enlarged locally in a non-general proportion, and the schematic diagrams are only examples and should not limit the scope of protection of the present invention herein. In addition, in actual production, three-dimensional spatial dimensions including length, width, and depth should be included.

[0040] At the same time, in the description of the present invention, it should be noted that the orientation or positional relationship indicated by terms such as "upper, lower, inner, and outer" is based on the orientation or positional relationship shown in the drawings, and is only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of the present invention. In addition, the terms "first, second, or third" are only used for descriptive purposes and cannot be understood as indicating or implying relative importance.

[0041] Unless otherwise clearly defined and limited in the present invention, the terms "mounted, connected, and coupled" should be understood in a broad sense. For example, it can be a fixed connection, a detachable connection, or an integral connection; it can also be a mechanical connection, an electrical connection, or a direct connection, and can also be indirectly connected through an intermediate medium, or can be the communication inside two elements. For those of ordinary skill in the art, the specific meanings of the above terms in the present invention can be understood according to specific situations.

[0042] Embodiment 1

[0043] Referring to Figure 1 and Figure 2 , which is the first embodiment of the present invention, this embodiment provides a difficult airway assessment method based on an audio and video feature large model, including:

[0044] S1. Real-time collect the difficult airway state features through an airway audio collection system and an airway video collection system, and transmit and process the collected difficult airway state features by using a high-speed and low-latency data network communication method;

[0045] Further, the processing method is to perform normalization processing and out-of-threshold data elimination processing on the data, as well as processing on the fusion large model training data sequence of the spatial mixed sequence and the temporal mixed sequence of the difficult airway audio-visual feature data;

[0046] Furthermore, the difficult airway state features at least include the audio feature data sequence of the difficult airway and the video feature data sequence of the difficult airway;

[0047] Specifically, the processing steps of the fusion large model training data sequence of the spatial mixed sequence and the temporal mixed sequence of the difficult airway audio-visual feature data are as follows:

[0048] S101. Calculate the spatial mixed sequence of the difficult airway audio-visual feature data spatial mixed sequence;

[0049] S102. Calculate the fusion value of the difficult airway audio-visual feature data temporal mixed sequence, the difficult airway audio-visual feature data spatial mixed sequence, and the difficult airway audio-visual feature data temporal mixed sequence;

[0050] S103. Calculate the fusion value adjustment coefficient of the difficult airway audio-visual feature data spatial mixed sequence and the difficult airway audio-visual feature data temporal mixed sequence;

[0051] S104. Calculate the fusion large model training data sequence of the difficult airway audio-visual feature data spatial mixed sequence and the difficult airway audio-visual feature data temporal mixed sequence;

[0052] Specifically, the formula for calculating the spatial mixed sequence of the difficult airway audio-visual feature data spatial mixed sequence is expressed as:

[0053]

[0054] where, d t is the difficult airway audio-visual feature data spatial mixed sequence at the current sampling moment; s is the audio feature data sequence of the difficult airway; v is the video feature data sequence of the difficult airway; n is the length of the feature data sequence; a1 is the balance coefficient of the audio feature data sequence of the difficult airway; a2 is the balance coefficient of the video feature data sequence of the difficult airway; β1 is the harmonic intensity of the audio feature data sequence of the difficult airway; β2 is the harmonic intensity of the video feature data sequence of the difficult airway;

[0055] Furthermore, the constraint condition of the formula for calculating the spatial mixed sequence of the difficult airway audio-visual feature data spatial mixed sequence is expressed as:

[0056] a1 + a2 = 1

[0057] β1 + β2 < a1s + a2v

[0058] Specifically, the formula for calculating the time mixed sequence of difficult airway audio-visual feature data is expressed as:

[0059]

[0060] where g t is the time mixed sequence of difficult airway audio-visual feature data at the current sampling moment; s(t - nΔt) is the audio feature data sequence of the difficult airway translated by Δt; v(t - nΔt) is the video feature data sequence of the difficult airway translated by Δt; ω n is the weight value of the nth Δt difficult airway feature data sequence;

[0061] Specifically, the formula for calculating the fusion value of the spatial mixed sequence and the time mixed sequence of difficult airway audio-visual feature data is expressed as:

[0062]

[0063] where X t is the fusion value of the spatial mixed sequence and the time mixed sequence of difficult airway audio-visual feature data;

[0064] Specifically, the formula for calculating the adjustment coefficient of the fusion value of the spatial mixed sequence and the time mixed sequence of difficult airway audio-visual feature data is expressed as:

[0065]

[0066] where γ s is the variance of the audio feature data sequence of the difficult airway; γ v is the variance of the video feature data sequence of the difficult airway; s t is the adjustment coefficient of the fusion value of the spatial mixed sequence and the time mixed sequence of difficult airway audio-visual feature data; w t is the fusion large model training data sequence of the spatial mixed sequence and the time mixed sequence of difficult airway audio-visual feature data; k is the adjustment coefficient scaling factor of the fusion value of the spatial mixed sequence and the time mixed sequence of difficult airway audio-visual feature data, and its value range is in the interval of 0 to 1; η is the fusion large model training data sequence scaling factor of the spatial mixed sequence and the time mixed sequence of difficult airway audio-visual feature data, and its value range is in the interval of 0 to 1;

[0067] It should be noted that s t and X t are in a relevant association relationship, and will As the fusion large model training data sequence of the spatial mixed sequence of difficult airway audio-visual feature data and the temporal mixed sequence of difficult airway audio-visual feature data, the value of t decreases as it increases within the range of (0, 0.25), and X t considers γ s the variance of the audio feature data sequence of the difficult airway, X t considers γ v the variance of the video feature data sequence of the difficult airway. When their values increase, it can decrease;

[0068] S2. Build a difficult airway large model through the large model server, receive the difficult airway state features transmitted by the high-speed and low-latency data network communication method, and perform large model training on the difficult airway state features at the current moment based on the difficult airway large model to obtain the prediction evaluation of the difficult airway large model;

[0069] Furthermore, the large model server constructs the difficult airway large model based on the difficult airway state features collected in the past and the difficult airway state features collected currently;

[0070] Among them, the difficult airway large model is privately and locally deployed using an open-source large model framework;

[0071] It should be noted that private local deployment can ensure local data processing, avoid transmitting sensitive medical data to external servers, reduce the risk of data leakage, and protect the privacy of patients;

[0072] Furthermore, the difficult airway large model iterates the training accuracy of the difficult airway large model by comparing the actually collected difficult airway state features with the prediction evaluation results of the difficult airway large model;

[0073] It should be noted that the iterative training process reduces the need for human intervention, reduces the workload of manual adjustment and possible human errors, improves the automation degree and work efficiency of the overall system, and can quickly respond to clinical needs. Especially in emergency situations, it ensures that the model can be improved and optimized within an extremely short time;

[0074] Furthermore, the difficult airway large model uses a local large model neural network to calculate the change trend of the difficult airway data sequence features and evaluates the tracheal intubability and tracheal intubation safety of the actual difficult airway according to preset rules;

[0075] Specifically, the evaluation of the preset rules can be based on the patient's own difficult airway status, so as to enable personalized tracheal intubation for each patient, further improving the success rate and safety of tracheal intubation operations;

[0076] S3. According to the prediction evaluation, establish a large model for difficult airway assisted decision-making, and iteratively optimize the large model for difficult airway through a real-time backpropagation mechanism;

[0077] Furthermore, use adaptive particle swarm optimization decision-making as the difficult airway large model assisted decision-making;

[0078] Moreover, the adaptive particle swarm optimization decision-making obtains the operation safety score of the difficult airway according to the intubability and tracheal intubation safety of the actual difficult airway and the prediction evaluation results of the difficult airway large model.

[0079] It should be noted that the adaptive particle swarm optimization decision-making can reduce the complications caused by tracheal intubation operations on the basis of the success rate and safety of tracheal intubation operations; and by obtaining the operation safety score of the difficult airway, it can help doctors with different qualifications better understand and judge the specific situation of the difficult airway.

[0080] Embodiment 2

[0081] Refer to Figure 3 , which is the second embodiment of the present invention. This embodiment provides a method for evaluating difficult airways based on a large model of audio and video features, including: In this embodiment, through a simulation experiment, the method for evaluating difficult airways based on a large model of audio and video features is used to evaluate the intubability and safety of actual patients;

[0082] The equipment used in the simulation experiment includes an airway audio acquisition system, an airway video acquisition system, and a transmission and processing device with a high-speed and low-latency data network communication method; the large model server is privately and locally deployed using an open-source large model framework, and the local large model neural network is used to calculate the change trend of the difficult airway data sequence features;

[0083] Ten patients with difficult airways were selected for data collection. Before tracheal intubation for each patient, the airway state characteristic data, including the audio feature data sequence and the video feature data sequence, were collected in real time using the airway audio acquisition system and the airway video acquisition system. First, the collected audio and video data were first normalized to eliminate individual differences between different patients. Then, the data were processed to remove out-of-threshold data to remove noise and outliers. The spatial mixing sequence and the temporal mixing sequence were calculated for the processed data respectively, and merged into the large model training data sequence. Secondly, the large model was trained using the large model server. The server trained the data using a neural network based on historical collected data and current real-time data. During the training process, the model parameters were optimized through a real-time backpropagation mechanism to continuously improve the prediction accuracy of the model. Finally, after the model training was completed, the prediction evaluation results of the difficult airway large model were obtained. The difficult airway large model evaluated the intubability and tracheal intubation safety of the actual difficult airway through an adaptive particle swarm optimization (APSO) decision. The operation safety score for each patient was calculated through APSO, providing auxiliary decision-making support for clinicians.

[0084] The method of the present invention was compared with the traditional subjective experience evaluation method, the single audio analysis evaluation method, and the single video analysis evaluation method, and the performance of each method in terms of the success rate and safety of tracheal intubation was evaluated. The simulated experimental results are shown in Tables 1 and 2.

[0085] Table 1 Method of the present invention

[0086]

[0087] Table 2 Traditional methods

[0088]

[0089] From Tables 1 and 2, it can be seen that the operation safety scores of the method of the present invention are 95, 90, 92, 97, 91, 86, 89, 93, 90, and 87 respectively. Compared with the traditional methods (85, 80, 82, 87, 81, 76, 79, 83, 80, and 77 respectively), the single audio analysis evaluation method (88, 83, 85, 90, 84, 79, 82, 86, 83, and 80 respectively), and the single video analysis evaluation method (90, 85, 87, 92, 86, 81, 84, 88, 85, and 82 respectively), the method of the present invention shows obvious advantages in the scores of each patient.

[0090] Moreover, through Figure 3It can be seen that the reduction in the variance of the fusion value reflects the improvement in the stability of the data sequence after adjustment. The smaller the variance, the smaller the data fluctuation and the better the stability of the system, demonstrating that after adjusting the data sequence by the method of the present invention, the abnormal fluctuations are significantly reduced, and the stability and reliability of the model training data are improved;

[0091] Through the comparison of simulation experiment data, it can be found that the difficult airway assessment method based on the large model of audio and video features is significantly higher than the traditional method and the single-dimensional analysis method in terms of operation safety score; the traditional method relies on the subjective experience of doctors and single-dimensional data, and is easily affected by personal experience and data noise, and the evaluation results have great subjectivity and uncertainty; while the method of the present invention combines audio and video feature data, and through the comprehensive analysis of multi-dimensional data and adaptive particle swarm optimization decision-making, can provide more comprehensive, scientific and accurate evaluation results.

[0092] Those skilled in the art should understand that the embodiments of the present invention can be provided as a method, a system, or a computer program product. Therefore, the present application can take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present application can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk memory, CD-ROM, optical memory, etc.) containing computer-usable program code. The solutions in the embodiments of the present application can be implemented in various computer languages, for example, object-oriented programming languages such as Java and interpreted scripting languages such as JavaScript.

[0093] The present application is described with reference to the flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each flow and / or block in the flowchart and / or block diagram, and the combination of flows and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to generate a machine, such that the instructions executed by the processor of the computer or other programmable data processing devices generate means for implementing the functions specified in Figure 1 one process or multiple processes and / or blocks Figure 1 one block or multiple blocks.

[0094] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to work in a specific manner, such that the instructions stored in the computer-readable memory generate a manufactured article including instruction means, and the instruction means implements the functions specified in Figure 1 one process or multiple processes and / or blocks Figure 1The functions specified in one or more boxes.

[0095] These computer program instructions can also be loaded onto a computer or other programmable data processing device, so that a series of operation steps are executed on the computer or other programmable device to generate a computer-implemented process. Thus, the instructions executed on the computer or other programmable device provide for implementing the steps of the functions specified in Figure 1 one process or more processes and / or boxes Figure 1 the functions specified in one box or more boxes.

[0096] Although the preferred embodiments of the present application have been described, those skilled in the art can make additional changes and modifications to these embodiments once they learn the basic creative concept. Therefore, the appended claims are intended to be construed as including the preferred embodiments and all changes and modifications that fall within the scope of the present application.

[0097] Obviously, those skilled in the art can make various changes and modifications to the present application without departing from the spirit and scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the claims of the present application and their equivalent technologies, the present application is also intended to include these modifications and variations.

Claims

1. A difficult airway assessment method based on a large model of audio and video features, characterized in that Including: Collect the difficult airway state characteristics in real time through the airway audio acquisition system and the airway video acquisition system, and transmit and process the collected difficult airway state characteristics by means of high-speed and low-latency data network communication; The real-time collection of the difficult airway state characteristics through the airway audio acquisition system and the airway video acquisition system, and the transmission and processing of the collected difficult airway state characteristics by means of high-speed and low-latency data network communication include: The processing method is to perform normalization processing and out-of-threshold data elimination processing on the data, and perform fusion large model training data sequence processing on the spatial mixed sequence and the temporal mixed sequence of the difficult airway audio-visual feature data; The difficult airway state characteristics at least include the audio feature data sequence of the difficult airway and the video feature data sequence of the difficult airway; The fusion large model training data sequence processing of the spatial mixed sequence and the temporal mixed sequence of the difficult airway audio-visual feature data includes: Calculate the spatial mixed sequence of the difficult airway audio-visual feature data; Specifically, the formula for calculating the spatial mixed sequence of the difficult airway audio-visual feature data is expressed as: Among them, d t is the spatial mixed sequence of difficult airway audio-video feature data at the current sampling moment; s is the audio feature data sequence of the difficult airway; v is the video feature data sequence of the difficult airway; n is the length of the feature data sequence; a1 is the balance coefficient of the audio feature data sequence of the difficult airway; a2 is the balance coefficient of the video feature data sequence of the difficult airway; β1 is the harmonic intensity of the audio feature data sequence of the difficult airway; β2 is the harmonic intensity of the video feature data sequence of the difficult airway; Calculate the fusion value of the temporal mixed sequence of the difficult airway audio-visual feature data, the spatial mixed sequence of the difficult airway audio-visual feature data, and the temporal mixed sequence of the difficult airway audio-visual feature data; Specifically, the formula for calculating the fusion value of the spatial mixed sequence and the temporal mixed sequence of the difficult airway audio-visual feature data is expressed as: Among them, X t is the fusion value of the spatial mixed sequence and the temporal mixed sequence of the difficult airway audio-video feature data; g t is the temporal mixed sequence of the difficult airway audio-video feature data at the current sampling moment; Also including: Calculate the fusion value adjustment coefficient of the spatial mixed sequence and the temporal mixed sequence of the difficult airway audio-visual feature data; Calculate the fusion large model training data sequence of the spatial mixed sequence and the temporal mixed sequence of the difficult airway audio-visual feature data; Build a difficult airway large model through the large model server, receive the difficult airway state characteristics transmitted by the high-speed and low-latency data network communication method, and perform current moment large model training on the difficult airway state characteristics based on the difficult airway large model to obtain the prediction evaluation of the difficult airway large model; According to the prediction evaluation, establish an auxiliary decision for the difficult airway large model, and perform iterative optimization on the difficult airway large model through the real-time backpropagation mechanism.

2. The difficult airway assessment method based on the large model of audio and video features according to claim 1, wherein Build a difficult airway large model through the large model server, and receive the difficult airway state characteristics transmitted by the high-speed and low-latency data network communication method, including: The large model server constructs the difficult airway large model based on the difficult airway state characteristics collected in the past and the difficult airway state characteristics collected currently; Among them, the difficult airway large model is privately and locally deployed using an open-source large model framework.

3. The difficult airway assessment method based on the large model of audio and video features according to claim 2, wherein Performing current moment large model training on the difficult airway state characteristics based on the difficult airway large model to obtain the prediction evaluation of the difficult airway large model includes: The difficult airway large model iteratively improves the training accuracy of the difficult airway large model by comparing the actually collected difficult airway state characteristics with the prediction evaluation results of the difficult airway large model.

4. The difficult airway assessment method based on the large model of audio and video features according to claim 3, characterized in that Also including: The large model for difficult airways uses a localized large model neural network to calculate the changing trend of the feature of the difficult airway data sequence, and evaluates the intubability and intubation safety of the actual difficult airway according to a preset rule.

5. The difficult airway assessment method based on the large model of audio and video features according to claim 3 or 4, characterized in that Based on the predicted evaluation, an auxiliary decision of the large model for difficult airways is established, including: Using the adaptive particle swarm optimization decision as the auxiliary decision of the large model for difficult airways; The adaptive particle swarm optimization decision obtains the operation safety score of the difficult airway according to the intubability and intubation safety of the actual difficult airway and the predicted evaluation result of the large model for difficult airways.

Images