Airway obstruction severity assessment method and system

Abstract

The invention belongs to the technical field of medical detection, and particularly relates to an airway obstruction severity assessment method and system. A medical data acquisition module is used for collecting multi-source medical data of a patient; the risk detection and evaluation module is used for inputting the preprocessed medical data into an airway obstruction AI evaluation model, and analyzing the preprocessed medical data based on the airway obstruction AI evaluation model to obtain an airway obstruction risk score of the patient; and the early warning evaluation module takes the risk score output by the result of the risk detection evaluation module as an object, carries out analysis from the early warning degree and the aging degree, obtains an early warning coefficient and an aging coefficient through analysis, and carries out data processing on the early warning coefficient and the aging coefficient to obtain an early warning evaluation value. The risk score output by the result of the risk detection and evaluation module is comprehensively analyzed from the risk degree and the processing degree of the time dimension, and the airway obstruction severity of the current patient can be evaluated more comprehensively and accurately.

Description

Technical Field

[0001] The present invention relates to the technical field of medical detection, and specifically to a method and system for evaluating the severity of airway obstruction. Background Art

[0002] Airway obstruction refers to partial or complete blockage of the airway, resulting in the inability of gas to pass through normally, thus causing symptoms such as dyspnea. Its causes are diverse and may include: upper airway obstruction: such as posterior tongue prolapse, nasal tumors, pharyngolaryngitis, and enlargement of organs in the throat (including lymph node enlargement) compressing the trachea or bronchus. Lower airway obstruction: including tracheal tumors, tracheal tuberculosis, tracheal foreign bodies, and inflammation, tuberculosis, and lymph node enlargement around the trachea or bronchus compressing the airway;

[0003] Currently, the assessment of the severity of airway obstruction mainly relies on clinical manifestations, laboratory tests, and imaging examinations. It is difficult to comprehensively integrate multi-source medical data and automatically judge the airway obstruction risk of patients. It is even more difficult to analyze the early warning level based on the severity assessment in multiple dimensions;

[0004] In view of the above technical deficiencies, a solution is proposed. Summary of the Invention

[0005] The purpose of the present invention is to provide a method and system for evaluating the severity of airway obstruction, which solves the problems that it is difficult to comprehensively integrate multi-source medical data and automatically judge the airway obstruction risk of patients in the prior art, and it is even more difficult to analyze the early warning level based on the severity assessment in multiple dimensions.

[0006] To achieve the above purpose, the present invention provides the following technical solutions:

[0007] A system for evaluating the severity of airway obstruction, comprising:

[0008] A medical data collection module: used to collect multi-source medical data of patients;

[0009] A risk detection and assessment module: input the preprocessed medical data into the airway obstruction AI assessment model, and analyze the preprocessed medical data based on the airway obstruction AI assessment model to obtain the airway obstruction risk score of the patient;

[0010] An early warning assessment module: taking the risk score output by the risk detection and assessment module as the object, analyzing from the aspects of early warning level and timeliness level, obtaining the early warning coefficient and timeliness coefficient, and performing data processing on the early warning coefficient and timeliness coefficient to obtain the early warning assessment value.

[0011] A further technical solution of the present invention: The specific operation process of the risk detection and assessment module is as follows:

[0012] AI Evaluation Model Loading: Load a pre-trained AI evaluation model for airway obstruction, where the AI evaluation model for airway obstruction is a deep learning network or a machine learning model;

[0013] Data Input: Input the preprocessed medical data into the AI evaluation model for airway obstruction;

[0014] Model Inference: The loaded AI evaluation model for airway obstruction performs inference calculations on the input medical data and obtains one or more predicted values related to the risk of airway obstruction;

[0015] Result Output: Output the airway obstruction risk assessment result of the AI evaluation model for airway obstruction in an understandable form, and output some additional information.

[0016] Further Technical Solution of the Present Invention: Result Explanation and Feedback: Provide an explanation of the airway obstruction risk assessment result through visualization technology or natural language processing technology.

[0017] Further Technical Solution of the Present Invention: Risk Management Recommendation: Generate personalized risk management recommendations based on the airway obstruction risk assessment result.

[0018] Further Technical Solution of the Present Invention: The analysis process of the warning coefficient is as follows:

[0019] Obtain the risk score output by the risk detection and assessment module, compare the risk score with the airway obstruction severity range, and determine the severity level and severity interval;

[0020] Perform a ratio process on the risk score and the maximum endpoint value of the corresponding severity interval to obtain the warning coefficient.

[0021] Further Technical Solution of the Present Invention: The airway obstruction severity range includes severity levels, specifically mild obstruction, moderate obstruction, and severe obstruction, and also includes severity intervals, specifically [A1, A2), [A2, A3), [A3, A4]; where [A1, A2) corresponds to mild obstruction, [A2, A3) corresponds to mild to moderate obstruction, and [A3, A4] corresponds to severe obstruction.

[0022] Further Technical Solution of the Present Invention: The analysis process of the timeliness coefficient is as follows:

[0023] Obtain the standard evaluation duration and standard mechanism duration corresponding to each severity level, calculate the ratio of the evaluation duration to the standard evaluation duration to obtain the evaluation coefficient, calculate the ratio of the mechanism duration to the standard mechanism duration to obtain the mechanism coefficient, and perform a weighting process on the evaluation coefficient and the mechanism coefficient to obtain the timeliness coefficient.

[0024] Further Technical Solution of the Present Invention: The process of obtaining the evaluation duration is as follows:

[0025] Obtain the time when the risk detection and evaluation module outputs the result, and the time when the medical data collection module starts to collect multi-source medical data of the patient, which are respectively recorded as the evaluation end time and the evaluation start time; calculate the time interval length between the evaluation end time and the evaluation start time to obtain the evaluation duration.

[0026] A further technical solution of the present invention: The process of obtaining the mechanism duration is as follows:

[0027] The time when the risk detection and evaluation module outputs the result, and the time when the patient to be diagnosed and treated is obtained, are respectively recorded as the evaluation end time and the mechanism start time, and calculate the time interval length between the mechanism start time and the evaluation end time to obtain the mechanism duration.

[0028] A method for evaluating the severity of airway obstruction, which uses the airway obstruction severity evaluation system as described above.

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

[0030] 1. In the present invention, by using AI technology to achieve accurate assessment and prediction of airway obstruction risk, the accuracy of diagnosis and the early detection rate are improved, and by integrating multi-source medical data to more comprehensively understand the patient's condition and risk factors, a more targeted treatment plan is provided, and it has the advantages of automation and intelligence, which can significantly reduce the doctor's workload and improve medical efficiency;

[0031] 2. Through the early warning evaluation module, the present invention comprehensively analyzes the risk score output by the risk detection and evaluation module from the risk degree itself and the processing degree in the time dimension. It can not only realize the early warning analysis and judgment of the severity of airway obstruction, the greater the early warning evaluation value, the higher the risk degree of subsequent diagnosis and treatment, but also realize the evaluation of the severity of airway obstruction of the current patient more comprehensively and accurately through the proximity of risk scores, the evaluation and treatment timeliness before and after scoring, and the obtained early warning evaluation value. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] For the convenience of those skilled in the art to understand, the present invention will be further described below with reference to the accompanying drawings;

[0033] Figure 1 It is the system block diagram of the first embodiment in the present invention;

[0034] Figure 2 It is the system block diagram of the third and fourth embodiments in the present invention;

[0035] Figure 3 It is the method flowchart of the second embodiment in the present invention;

[0036] Figure 4 This is the flowchart of the method in the fifth embodiment of the present invention. Specific implementation manners

[0037] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the 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.

[0038] Embodiment 1: As Figure 1 shown, a system for evaluating the severity of airway obstruction proposed by the present invention includes a server, a medical data collection module, a medical data preprocessing module, a risk detection and evaluation module, and a result output module;

[0039] The medical data collection module is used to collect multi-source medical data of patients, including but not limited to the medical history, genetic information, biochemical marker information, imaging examination information, etc. of the patients, and send the multi-source medical data of the patients to the medical data preprocessing module through the server; it should be noted that the multi-source medical data of the patients can be collected through channels such as the electronic medical record system, the gene detection platform, and the laboratory information management system;

[0040] The medical data preprocessing module performs preprocessing operations on the received medical data, including operations such as cleaning, sorting, and standardization, to ensure data quality and consistency, and sends the preprocessed medical data to the risk detection and evaluation module through the server;

[0041] The risk detection and evaluation module inputs the preprocessed medical data into the airway obstruction AI evaluation model, analyzes the preprocessed medical data based on the airway obstruction AI evaluation model, and obtains the airway obstruction risk assessment result of the patient; it should be noted that deep learning algorithms (such as convolutional neural networks, recurrent neural networks, etc.) and machine learning algorithms (such as support vector machines, random forests, etc.) are used to construct a prediction model for airway obstruction AI. During the model training process, methods such as cross-validation are used for model selection and parameter optimization to improve the generalization ability and prediction accuracy of the model;

[0042] The risk detection and evaluation module sends the airway obstruction risk assessment result of the patient to the result output module through the server. Among them, the evaluation result can be output in the form of a text report, a chart, etc. as needed; the result output module displays the airway obstruction risk assessment result of the patient; the specific operation process of the risk detection and evaluation module is as follows:

[0043] AI Evaluation Model Loading: Load a pre-trained AI evaluation model for airway obstruction. The airway obstruction AI evaluation model can be a deep learning network (such as a convolutional neural network, a recurrent neural network, or a fully connected neural network) or a machine learning model (such as a random forest, a support vector machine, etc.). During the training process, the airway obstruction AI evaluation model has learned how to extract features related to the risk of airway obstruction from the input data and establish complex relationships between these features and the risk of airway obstruction;

[0044] Data Input: Input the preprocessed medical data into the airway obstruction AI evaluation model. It should be noted that since the original data may contain a large amount of redundant information or information not directly related to the RA risk, a feature extraction step is performed before inputting the preprocessed medical data into the airway obstruction AI evaluation model, enabling the airway obstruction AI evaluation model to focus on the features most important for airway obstruction risk assessment. Feature extraction can be achieved through various methods, such as statistical-based methods, machine learning-based feature selection algorithms, or deep learning-based automatic feature learning, etc.;

[0045] Model Inference: The loaded airway obstruction AI evaluation model performs inference calculations on the input medical data and obtains one or more prediction values related to the risk of airway obstruction. These prediction values represent specific indicators such as the probability of a patient having airway obstruction, the risk level, or the predicted incidence rate, etc.;

[0046] Result Output: Output the airway obstruction risk assessment result of the airway obstruction AI evaluation model in an understandable form, such as a numerical risk score, a risk level (such as mild obstruction, moderate obstruction, severe obstruction), or a predicted incidence rate range, etc.; and output some additional information, such as the confidence in the prediction result and the risk factor analysis information related to the prediction result, etc.;

[0047] Result Interpretation and Feedback: Provide an interpretation of the airway obstruction risk assessment result, achieved through visualization techniques (such as charts, heatmaps, etc.) or natural language processing techniques (such as generating explanatory text);

[0048] Risk Management Recommendations: Based on the airway obstruction risk assessment result, generate personalized risk management recommendations, such as suggesting that the patient undergo further medical examinations, recommending specific treatment plans, or providing guidance on lifestyle and health management, etc.

[0049] The present invention uses AI technology to accurately evaluate and predict the risk of airway obstruction, improving the accuracy of diagnosis and the early detection rate. By integrating multi-source medical data, it can comprehensively understand the patient's condition and risk factors, provide a more targeted treatment plan, and the system has the characteristics of automation and intelligence, which can reduce the workload of doctors and improve medical efficiency.

[0050] Embodiment 2: As Figure 3 shown, a method and system for evaluating the severity of airway obstruction proposed by the present invention further includes:

[0051] Early warning evaluation module: Taking the risk score output by the risk detection and evaluation module as the object, analyzing from the aspects of early warning degree and timeliness degree, obtaining an early warning coefficient and a timeliness coefficient, and performing data processing on the early warning coefficient and the timeliness coefficient to obtain an early warning evaluation value; The specific operation process of this early warning evaluation module is as follows:

[0052] Obtain the risk score output by the risk detection and evaluation module, compare the risk score with the severe range of airway obstruction, and determine the severity level and severity interval; Specifically, the severe range of airway obstruction includes severity levels, specifically mild obstruction, moderate obstruction, and severe obstruction, and also includes severity intervals, specifically [A1, A2), [A2, A3), [A3, A4]; Among them, [A1, A2) corresponds to mild obstruction, [A2, A3) corresponds to mild to moderate obstruction, and [A3, A4] corresponds to severe obstruction;

[0053] Perform a ratio process on the risk score and the maximum endpoint value of the corresponding severity interval to obtain an early warning coefficient; This early warning coefficient reflects the proximity of the change in the severity level of the current airway obstruction patient. The larger the early warning coefficient, the higher the risk level at the current level;

[0054] Obtain the time output by the risk detection and evaluation module, and the time when the medical data collection module starts to collect multi-source medical data of the patient, which are respectively recorded as the evaluation end time and the evaluation start time; Calculate the time interval length between the evaluation end time and the evaluation start time to obtain the evaluation duration;

[0055] Again, the time output by the risk detection and evaluation module, and the time for treating the scored patient (the time for treatment is the time for arranging the patient for medication and surgery), are respectively recorded as the evaluation end time and the mechanism start time, and calculate the time interval length between the mechanism start time and the evaluation end time to obtain the mechanism duration;

[0056] Obtain the standard evaluation duration and the standard mechanism duration corresponding to each severity level. Calculate the ratio of the evaluation duration to the standard evaluation duration to obtain an evaluation coefficient, and calculate the ratio of the mechanism duration to the standard mechanism duration to obtain a mechanism coefficient. Perform weight processing on the evaluation coefficient and the mechanism coefficient to obtain a timeliness coefficient. Preferably, the weight factors of the evaluation coefficient and the mechanism coefficient can be 0.19 and 0.81 respectively; this timeliness coefficient reflects, with the time of the risk score as the demarcation point, a data representation of the evaluation efficiency of the airway obstruction severity and the efficiency of matching the diagnosis and treatment mechanism. The smaller the timeliness coefficient, the faster and more effectively the current patient can be evaluated, and the faster and more effectively treatment can be carried out after evaluation;

[0057] Among them, the standard evaluation duration and the standard mechanism duration are obtained by technicians based on historical data statistics;

[0058] Through the following formula, the early warning evaluation value is calculated, and the formula is as follows:

[0059] ZP = XY * XT -1

[0060] Among them, ZP is the early warning evaluation value, XY is the early warning coefficient, and XT is the timeliness coefficient;

[0061] In this embodiment, through the early warning evaluation module, a comprehensive analysis is carried out based on the risk score output by the risk detection and evaluation module from the aspects of its own risk level and the processing degree in the time dimension. It can not only realize the early warning analysis and judgment of the airway obstruction severity, and the larger the early warning evaluation value, the higher the risk degree of the subsequent diagnosis and treatment. It can also realize the early warning evaluation value obtained through the closeness of the risk score, the evaluation and treatment timeliness before and after the score, and can more comprehensively and accurately evaluate the airway obstruction severity of the current patient.

[0062] Embodiment 3: As Figure 2 shown, the difference between this embodiment and Embodiment 1 is that the server is communicatively connected to the optimization and upgrade early warning module, and the optimization and upgrade early warning module is used to set the detection period. Preferably, the detection period is sixty days; analyze the performance of the airway obstruction risk assessment within the detection period, and judge whether to generate an optimization and upgrade early warning signal through the analysis;

[0063] And send the optimization and upgrade early warning signal to the management terminal through the server. When the management terminal receives the optimization and upgrade early warning signal, it issues a corresponding warning to remind the system operation and maintenance management personnel to optimize and upgrade the system as needed, so as to ensure the accuracy of the subsequent risk assessment results for airway obstruction and the analysis and processing efficiency; the specific analysis process of the optimization and upgrade early warning module is as follows:

[0064] All airway obstruction risk assessment results within the detection period are obtained. If it is determined that the airway obstruction risk of the corresponding patient is high but the corresponding patient is not diagnosed with airway obstruction after diagnosis, or if it is determined that the airway obstruction risk of the corresponding patient is low but the corresponding patient is diagnosed with airway obstruction after diagnosis, it indicates that the accuracy of the corresponding assessment result is poor, and the corresponding airway obstruction risk assessment result is marked as a risk misassessment result;

[0065] The number of risk misassessment results within the detection period is obtained and its ratio is calculated with the total number of airway obstruction risk assessment results to obtain a risk misassessment judgment value. Among them, the larger the value of the risk misassessment judgment value, the worse the accuracy of the risk assessment results of the system for airway obstruction within the detection period; The risk misassessment judgment value is numerically compared with a preset risk misassessment judgment threshold. If the risk misassessment judgment value exceeds the preset risk misassessment judgment threshold, it indicates that the accuracy of the risk assessment results of the system for airway obstruction within the detection period is poor, and an optimization and upgrade warning signal is generated.

[0066] Furthermore, if the risk misassessment judgment value does not exceed the preset risk misassessment judgment threshold, the analysis duration corresponding to the corresponding airway obstruction risk assessment result is collected (i.e., the interval duration between the output time of the risk assessment result and the collection time of the medical data). The analysis duration is numerically compared with a preset analysis duration threshold. If the analysis duration exceeds the preset analysis duration threshold, it indicates that the processing and analysis efficiency of the corresponding airway obstruction risk assessment is low, and the corresponding analysis duration is marked as an abnormal analysis duration;

[0067] The number of abnormal analysis durations within the detection period is obtained and its ratio is calculated with the total number of airway obstruction risk assessment results to obtain an abnormal analysis situation value, and the average value of all analysis durations within the detection period is calculated to obtain an analysis efficiency situation value;

[0068] The risk misassessment judgment value WF, the abnormal analysis situation value WK, and the analysis efficiency situation value WN are numerically calculated through the formula WP = ry1 * WF + ry2 * WK + ry3 * WN / (ry1 + ry2) to obtain an optimization and upgrade warning value WP, where ry1, ry2, and ry3 are preset proportionality coefficients, and ry1 > ry2 > ry3 > 0; Moreover, the larger the value of the optimization and upgrade warning value WP, the worse the overall assessment performance of the system within the detection period;

[0069] The above formula is dimensionless and takes its numerical calculation. The formula is obtained by collecting a large amount of data for software simulation to obtain a formula closest to the actual situation. The preset parameters in the formula are set by those skilled in the art according to the actual situation; The optimization and upgrade warning value WP is numerically compared with a preset optimization and upgrade warning threshold. If the optimization and upgrade warning value WP exceeds the preset optimization and upgrade warning threshold, it indicates that the overall assessment performance of the system within the detection period is poor, and an optimization and upgrade warning signal is generated.

[0070] Example 4: As Figure 2 shown, the difference between this embodiment and Embodiment 1 and Embodiment 2 is that the server is communicatively connected to the verification and analysis module. If no optimization and upgrade warning signal is generated, several groups of system evaluation tests are performed through the verification and analysis module, and based on the test results, it is judged whether to generate an evaluation instability signal, and the evaluation instability signal is sent to the management terminal through the server;

[0071] When the management terminal receives the evaluation instability signal, it issues a corresponding warning, which can accurately feedback the result stability and efficiency stability of the system evaluation, and timely remind the system operation and maintenance management personnel to check and optimize the system as needed, so as to ensure the accuracy and stability of the subsequent risk assessment for airway obstruction; The specific test and analysis process of the verification and analysis module is as follows:

[0072] Obtain several groups of medical data packets prepared in advance and used for system evaluation tests. Preferably, the number of groups of medical data packets is not less than fifteen groups; send the corresponding medical data packets to the medical data acquisition module, and perform several airway obstruction risk assessments based on the corresponding medical data packets. Preferably, the number of tests for the corresponding medical data packets is not less than five times; if there are differences in the several risk assessment results for the corresponding medical data packets, it indicates that the test result stability for the corresponding medical data packets is poor, then assign a judgment symbol RP-1 to the corresponding medical data packet;

[0073] If there are no differences in the several risk assessment results for the corresponding medical data packets, it indicates that the test result stability for the corresponding medical data packets is good, then calculate the variance of the several analysis durations for the corresponding medical data packets to obtain an analysis time deviation value, and compare the analysis time deviation value with a preset analysis time deviation threshold. If the analysis time deviation value exceeds the preset analysis time deviation threshold, it indicates that the test efficiency stability for the corresponding medical data packets is poor, then assign a judgment symbol RP-1 to the corresponding medical data packet;

[0074] Obtain the number of medical data packets assigned with the judgment symbol RP-1 and mark it as the abnormal test count value, and calculate the ratio of the abnormal test count value to the number of groups of medical data packets to obtain the abnormal test occupancy value. Among them, the larger the value of the abnormal test occupancy value, the worse the overall system evaluation stability; compare the abnormal test occupancy value with a preset abnormal test occupancy threshold. If the abnormal test occupancy value exceeds the preset abnormal test occupancy threshold, it indicates that the overall system evaluation stability is poor, then generate an evaluation instability signal.

[0075] Example 5: As Figure 3 shown, a method for evaluating the severity of airway obstruction proposed by the present invention includes the following steps:

[0076] Step 1: The medical data acquisition module collects multi-source medical data of the patient;

[0077] Step 2: The medical data preprocessing module performs preprocessing operations on the collected medical data;

[0078] Step 3: The risk detection and assessment module inputs the preprocessed medical data into the airway obstruction AI assessment model;

[0079] Step 4: Analyze the preprocessed medical data based on the airway obstruction AI assessment model to obtain the airway obstruction risk assessment result of the patient;

[0080] Step 5: The result output module displays the airway obstruction risk assessment result of the patient.

[0081] The working principle of the present invention: When in use, the multi-source medical data of the patient is collected through the medical data collection module. The medical data preprocessing module performs preprocessing operations on the received medical data. The risk detection and assessment module inputs the preprocessed medical data into the airway obstruction AI assessment model. Analyze the preprocessed medical data based on the airway obstruction AI assessment model to obtain the airway obstruction risk assessment result of the patient. The result output module displays the airway obstruction risk assessment result of the patient. By using AI technology, the accurate assessment and prediction of airway obstruction risk are realized, the diagnostic accuracy and early detection rate are improved, and by integrating multi-source medical data, a more comprehensive understanding of the patient's condition and risk factors is achieved, providing a more targeted treatment plan, and having the advantages of automation and intelligence, which can significantly reduce the workload of doctors and improve medical efficiency; and by optimizing and upgrading the warning module, analyze the airway obstruction risk assessment performance within the detection period to determine whether to generate an optimized upgrade warning signal. If no upgrade and optimization warning signal is generated, perform several groups of system assessment tests through the verification analysis module and determine whether to generate an assessment instability signal. When an optimized upgrade warning signal or an assessment instability signal is generated, the management terminal issues a corresponding warning, which can timely remind the system operation and maintenance management personnel to take targeted improvement measures to ensure the accuracy, efficiency and stability of the subsequent risk assessment for airway obstruction.

[0082] The preferred embodiments of the present invention disclosed above are only used to help explain the present invention. The preferred embodiments do not describe all the details in detail, nor do they limit the present invention to only the specific implementation manners. Obviously, many modifications and variations can be made according to the content of this specification. This specification selects and specifically describes these embodiments to better explain the principle and practical application of the present invention, so that those skilled in the relevant technical field can well understand and utilize the present invention. The present invention is only limited by the claims and their full scope and equivalents.

Claims

1. A system for assessing the severity of airway obstruction, characterized in that: include: Medical data collection module: used to collect multi-source medical data of patients; Risk detection and assessment module: input the pre-processed medical data into the airway obstruction AI assessment model, analyze the pre-processed medical data based on the airway obstruction AI assessment model, and obtain the patient's airway obstruction risk score; Early warning assessment module: Taking the risk score output by the risk detection and assessment module as the object, it analyzes the warning degree and timeliness degree, obtains the warning coefficient and timeliness coefficient, and processes the warning coefficient and timeliness coefficient to obtain the early warning assessment value.

2. The airway obstruction severity assessment system according to claim 1, characterized in that: The specific operation process of the risk detection and assessment module is as follows: AI assessment model loading: load a pre-trained airway obstruction AI assessment model, where the airway obstruction AI assessment model is a deep learning network or a machine learning model; Data input: input the pre-processed medical data into the airway obstruction AI assessment model; Model reasoning: The loaded airway obstruction AI assessment model performs reasoning calculations on the input medical data and derives one or more predicted values ​​related to the risk of airway obstruction; Result output: The airway obstruction risk assessment results of the airway obstruction AI assessment model are output in an understandable form, as well as some additional information.

3. The airway obstruction severity assessment system according to claim 2, characterized in that: Result interpretation and feedback: Provide explanations of airway obstruction risk assessment results through visualization technology or natural language processing technology.

4. The airway obstruction severity assessment system according to claim 3, characterized in that: Risk management recommendations: Generate personalized risk management recommendations based on airway obstruction risk assessment results.

5. The airway obstruction severity assessment system according to claim 1, characterized in that: The analysis process of the early warning coefficient is: Obtain the risk score output by the risk detection and assessment module, compare the risk score with the severity range of airway obstruction, and determine the severity level and severity range; The risk score is compared with the maximum endpoint value of the corresponding severity interval to obtain the warning coefficient.

6. The airway obstruction severity assessment system according to claim 5, characterized in that: The severity range of airway obstruction includes severity levels, specifically mild obstruction, moderate obstruction, and severe obstruction, and also includes severity intervals, specifically [A1, A2), [A2, A3), [A3, A4]; among them, [A1, A2) corresponds to mild obstruction, [A2, A3) corresponds to mild to moderate obstruction, and [A3, A4] corresponds to severe obstruction.

7. The airway obstruction severity assessment system according to claim 5, characterized in that: The analysis process of the aging coefficient is: Obtain the standard assessment time and standard mechanism time corresponding to each severity level, calculate the ratio of the assessment time to the standard assessment time to obtain the assessment coefficient, calculate the ratio of the mechanism time to the standard mechanism time to obtain the mechanism coefficient, perform weight processing on the assessment coefficient and the mechanism coefficient to obtain the timeliness coefficient.

8. The airway obstruction severity assessment system according to claim 7, characterized in that: The process of obtaining the evaluation duration is as follows: The time when the risk detection and assessment module outputs the results, and the time when the medical data acquisition module starts collecting the patient's multi-source medical data, are recorded as the assessment end time and the assessment start time respectively; the time interval between the assessment end time and the assessment start time is calculated to obtain the assessment duration.

9. The airway obstruction severity assessment system according to claim 8, characterized in that: The process of obtaining the mechanism duration is as follows: The time when the risk detection and assessment module outputs the results, and the time when the patient with the score is required to be diagnosed and treated, are recorded as the assessment end time and the mechanism start time, respectively. The time interval between the mechanism start time and the assessment end time is calculated to obtain the mechanism duration.

10. A method for assessing the severity of airway obstruction, characterized in that: The method adopts the airway obstruction severity assessment system as described in any one of claims 1-9.

Images