Intelligent assistance system for lung function testing

Abstract

The present application discloses an intelligent auxiliary system for lung function testing, which relates to the field of intelligent monitoring. It collects the user's vital capacity time series data by integrating a peak flow meter and a spirometer, and uses a data processing and analysis algorithm based on artificial intelligence and deep learning to analyze the vital capacity time series data, so as to capture the frequency domain semantic time series aggregation features of the vital capacity after separating the intrinsic mode functions of different frequency components, so as to automatically perform the user's lung function test and give the test results of whether the lung function is abnormal. In this way, it is possible to realize the automatic and intelligent detection of the user's lung function status, so as to assist the user himself or medical personnel so that they can timely discover early or hidden lung function abnormalities, and provide strong support for clinical diagnosis and personal health management.

Description

Technical Field

[0001] The present application relates to the field of intelligent detection, and more specifically, to an intelligent auxiliary system for lung function detection. Background Art

[0002] With the development of society and the improvement of people's health awareness, lung function testing has gradually become an important part of daily health management. Lung function testing can help diagnose respiratory diseases such as asthma, chronic obstructive pulmonary disease (COPD), pulmonary fibrosis, etc., and is of great significance for the early detection and treatment of diseases.

[0003] Traditional pulmonary function testing methods mainly rely on the doctor's experience and manual operation, which is not only time-consuming and labor-intensive, but also its accuracy is affected by many factors, making it difficult to meet the fast and accurate medical needs of modern society. Specifically, traditional pulmonary function tests usually require professional medical personnel to operate and rely on the doctor's experience and judgment, which makes the test results susceptible to personal factors and leads to inconsistent results. Although existing pulmonary function testing equipment is already quite advanced, due to the lack of intelligent data processing and analysis methods, some subtle changes or early symptoms may not be captured in a timely and accurate manner, thus affecting the accuracy of the doctor's diagnosis.

[0004] Therefore, an intelligent assistance system for lung function testing is desired. Summary of the invention

[0005] In order to solve the above technical problems, this application is proposed.

[0006] The present application provides an intelligent auxiliary system for lung function testing, comprising:

[0007] A vital capacity data acquisition module, used for acquiring a time queue of vital capacity acquired by a peak flow meter and a spirometer;

[0008] A vital capacity data decomposition module, used for performing empirical mode decomposition on the time queue of the vital capacity to obtain a set of vital capacity intrinsic mode functions and a set of expiratory flow rate intrinsic mode functions;

[0009] A vital capacity intrinsic modal frequency domain semantic feature extraction module, used for extracting features from the set of vital capacity intrinsic modal functions to obtain a set of vital capacity intrinsic modal frequency domain semantic coding features;

[0010] A vital capacity decomposition kernel feature dynamic aggregation module is used to perform kernel feature dynamic aggregation analysis on the set of vital capacity intrinsic modal frequency domain semantic coding features to obtain vital capacity time series kernel features; wherein, the vital capacity decomposition kernel feature dynamic aggregation module includes: an intrinsic modal frequency domain semantic kernel factor calculation unit, which is used to calculate the static energy potential factor and dynamic energy potential factor of each vital capacity intrinsic modal frequency domain semantic coding feature in the set of vital capacity intrinsic modal frequency domain semantic coding features to obtain a set of vital capacity intrinsic modal frequency domain semantic static energy potential factors and a set of vital capacity intrinsic modal frequency domain semantic dynamic energy potential factors; a vital capacity time series kernel feature aggregation unit, which is used to perform kernel semantic aggregation on the set of vital capacity intrinsic modal frequency domain semantic coding features based on the set of vital capacity intrinsic modal frequency domain semantic static energy potential factors and the set of vital capacity intrinsic modal frequency domain semantic dynamic energy potential factors to obtain the vital capacity time series kernel features;

[0011] The pulmonary function detection module is used to perform pulmonary function detection based on the vital capacity time series kernel feature to determine the detection result.

[0012] Based on the above-mentioned intelligent auxiliary system for pulmonary function testing, the vital capacity intrinsic modal frequency domain semantic feature extraction module is used to: input the set of vital capacity intrinsic modal functions into the intrinsic modal feature extractor based on the 1D-CNN model to obtain a set of vital capacity intrinsic modal frequency domain semantic coding feature vectors as the set of vital capacity intrinsic modal frequency domain semantic coding features.

[0013] Based on the above-mentioned intelligent auxiliary system for pulmonary function testing, the intrinsic modal frequency domain semantic kernel factor calculation unit includes:

[0014] A mean calculation subunit, used for calculating the position mean vector of the set of the vital capacity intrinsic modal frequency domain semantic encoding feature vectors to obtain the vital capacity intrinsic modal frequency domain semantic initial aggregation representation vector;

[0015] A vital capacity intrinsic modal frequency domain semantic static energy potential factor calculation subunit, used to calculate the static energy potential factor of each vital capacity intrinsic modal frequency domain semantic coding feature vector in the set of vital capacity intrinsic modal frequency domain semantic coding feature vectors to obtain the set of vital capacity intrinsic modal frequency domain semantic static energy potential factors;

[0016] The vital capacity intrinsic modal frequency domain semantic dynamic energy potential factor calculation subunit is used to calculate the dynamic energy potential factor of each vital capacity intrinsic modal frequency domain semantic coding feature vector in the set of vital capacity intrinsic modal frequency domain semantic coding feature vectors relative to the vital capacity intrinsic modal frequency domain semantic initial aggregation representation vector to obtain the set of vital capacity intrinsic modal frequency domain semantic dynamic energy potential factors.

[0017] Based on the above-mentioned intelligent auxiliary system for lung function testing, the vital capacity intrinsic modal frequency domain semantic static energy potential factor calculation subunit is used to:

[0018] Calculating the mean and standard deviation of the vital capacity intrinsic modal frequency domain semantic encoding feature vector to obtain the vital capacity intrinsic modal frequency domain semantic mean and the vital capacity intrinsic modal frequency domain semantic standard deviation;

[0019] Calculating the position difference between the vital capacity intrinsic modal frequency domain semantic coding feature vector and the vital capacity intrinsic modal frequency domain semantic mean, and then dividing the difference by the vital capacity intrinsic modal frequency domain semantic standard deviation by position point to obtain the vital capacity intrinsic modal frequency domain semantic distribution representation vector;

[0020] After calculating the fourth power of the characteristic values ​​of each position in the vital capacity intrinsic modal frequency domain semantic distribution representation vector, the mean value of the fourth power modulation vector is calculated to obtain the vital capacity intrinsic modal frequency domain semantic static energy potential factor.

[0021] Based on the above-mentioned intelligent auxiliary system for lung function testing, the vital capacity intrinsic modal frequency domain semantic dynamic energy potential factor calculation subunit is used to:

[0022] Calculate the multiplication between the vital capacity intrinsic modal frequency domain semantic initial aggregation representation vector and the vital capacity intrinsic modal frequency domain semantic initial clustering weight matrix to obtain the vital capacity intrinsic modal frequency domain semantic initial aggregation representation modulation vector;

[0023] Calculating the multiplication between the vital capacity intrinsic modal frequency domain semantic coding feature vector and the vital capacity intrinsic modal frequency domain semantic weight matrix to obtain the vital capacity intrinsic modal frequency domain semantic modulation vector;

[0024] The vital capacity intrinsic modal frequency domain semantic initial aggregation representation modulation vector is multiplied by the vital capacity intrinsic modal frequency domain semantic modulation vector by position point, and then multiplied by the vital capacity intrinsic modal frequency domain semantic offset vector by position point to obtain the vital capacity intrinsic modal frequency domain semantic dynamic modulation representation vector;

[0025] The square of the Frobenius norm of the vital capacity intrinsic modal frequency domain semantic dynamic modulation representation vector is calculated to obtain the vital capacity intrinsic modal frequency domain semantic dynamic energy potential factor.

[0026] Based on the above-mentioned intelligent auxiliary system for lung function detection, the vital capacity time series kernel feature aggregation unit includes:

[0027] A vital capacity intrinsic modal frequency domain semantic superposition state energy potential factor calculation subunit, used to determine the superposition state energy potential factor of each vital capacity intrinsic modal frequency domain semantic coding feature vector in the set of vital capacity intrinsic modal frequency domain semantic coding feature vectors based on the vital capacity intrinsic modal frequency domain semantic static energy potential factor and the vital capacity intrinsic modal frequency domain semantic dynamic energy potential factor of each vital capacity intrinsic modal frequency domain semantic coding feature vector in the set of vital capacity intrinsic modal frequency domain semantic coding feature vectors to obtain a set of vital capacity intrinsic modal frequency domain semantic superposition state energy potential factors;

[0028] An energy potential factor gating weighting subunit is used to input the set of energy potential factors of the vital capacity intrinsic modal frequency domain semantic superposition state into an energy potential factor gating weighting network to obtain a set of vital capacity intrinsic modal frequency domain semantic feature superposition energy potential weight coefficients;

[0029] The vital capacity time series kernel feature calculation subunit is used to calculate the position-weighted sum of the set of vital capacity inherent modal frequency domain semantic coding feature vectors based on the set of potential weight coefficients superimposed on the vital capacity inherent modal frequency domain semantic features to obtain the vital capacity time series kernel feature vector as the vital capacity time series kernel feature.

[0030] Based on the above-mentioned intelligent auxiliary system for lung function testing, the vital capacity intrinsic modal frequency domain semantic superposition state energy potential factor calculation subunit is used to:

[0031] Calculate the natural exponential function value with the natural constant e as the base and the vital capacity intrinsic modal frequency domain semantic static energy potential factor of the vital capacity intrinsic modal frequency domain semantic encoding feature vector as the exponential power to obtain the vital capacity intrinsic modal frequency domain semantic static energy potential class support factor;

[0032] Calculate the natural exponential function value with the natural constant e as the base and the square of the vital capacity intrinsic modal frequency domain semantic dynamic energy potential factor of the vital capacity intrinsic modal frequency domain semantic encoding feature vector as the exponential power to obtain the vital capacity intrinsic modal frequency domain semantic dynamic energy potential class support factor;

[0033] The sum of the vital capacity intrinsic modal frequency domain semantic static energy potential class support factor and the vital capacity intrinsic modal frequency domain semantic dynamic energy potential class support factor is calculated to obtain the vital capacity intrinsic modal frequency domain semantic superposition state energy potential factor.

[0034] Based on the above-mentioned intelligent auxiliary system for pulmonary function testing, the energy potential factor gating weighting subunit includes:

[0035] The secondary subunit for calculating the vital capacity intrinsic modal frequency domain semantic superposition state energy potential class support factor is used to calculate the natural exponential function value with the natural constant e as the base and the negative number of the vital capacity intrinsic modal frequency domain semantic superposition state energy potential factor as the exponential power to obtain the vital capacity intrinsic modal frequency domain semantic superposition state energy potential class support factor;

[0036] A secondary subunit for calculating the normalized representation coefficient of the superposition energy potential of the intrinsic modal frequency domain semantic features of vital capacity, which is used to calculate the inverse of the sum of the intrinsic modal frequency domain semantic superposition state energy potential class support factor of the vital capacity and a constant one to obtain the normalized representation coefficient of the superposition energy potential of the intrinsic modal frequency domain semantic features of vital capacity;

[0037] A masked secondary subunit is used to perform masking processing on the normalized representation coefficient of the vital capacity intrinsic modal frequency domain semantic feature superposition energy potential to obtain a vital capacity intrinsic modal frequency domain semantic feature superposition energy potential weight coefficient;

[0038] Among them, the masked secondary sub-unit is further used to: in response to the normalized representation coefficient of the superposition potential of the inherent modal frequency domain semantic features of the vital capacity being greater than a predetermined threshold, take the normalized representation coefficient of the superposition potential of the inherent modal frequency domain semantic features of the vital capacity as the original value, otherwise set it to 0, so as to obtain the superposition potential weight coefficient of the inherent modal frequency domain semantic features of the vital capacity.

[0039] Based on the above-mentioned intelligent auxiliary system for pulmonary function testing, the pulmonary function testing module is used to: input the vital capacity time series kernel feature vector into the classifier-based pulmonary function testing module to obtain a test result, and the test result is used to indicate whether there is an abnormality.

[0040] Compared with the prior art, the intelligent auxiliary system for lung function testing provided by the present application collects the user's vital capacity time series data by integrating a peak flow meter and a spirometer, and uses a data processing and analysis algorithm based on artificial intelligence and deep learning to analyze the vital capacity time series data, thereby capturing the frequency domain semantic time series aggregation features of the vital capacity after separating the intrinsic mode functions of different frequency components, so as to automatically perform the user's lung function test and provide the test results of whether there is any abnormality in the lung function. In this way, it is possible to realize the automated and intelligent detection of the user's lung function status, so as to assist the user himself or medical personnel so that they can promptly discover early or hidden lung function abnormalities, and provide strong support for clinical diagnosis and personal health management. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] By describing the embodiments of the present application in more detail in conjunction with the accompanying drawings, the above and other purposes, features and advantages of the present application will become more apparent. The accompanying drawings are used to provide a further understanding of the embodiments of the present application and constitute a part of the specification. Together with the embodiments of the present application, they are used to explain the present application and do not constitute a limitation of the present application. In the accompanying drawings, the same reference numerals generally represent the same components or steps.

[0042] Figure 1 is a block diagram of an intelligent auxiliary system for lung function testing according to an embodiment of the present application;

[0043] Figure 2 A schematic diagram of data flow of an intelligent auxiliary system for lung function testing according to an embodiment of the present application;

[0044] Figure 3 It is a block diagram of a vital capacity decomposition kernel feature dynamic aggregation module in an intelligent assistance system for lung function testing according to an embodiment of the present application. DETAILED DESCRIPTION

[0045] Below, the exemplary embodiments according to the present application will be described in detail with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments of the present application, and it should be understood that the present application is not limited to the exemplary embodiments described here.

[0046] As shown in this application and claims, unless the context clearly indicates an exception, the words "a", "an", "an" and / or "the" do not refer to the singular and may also include the plural. Generally speaking, the terms "include" and "comprise" only indicate the inclusion of the steps and elements that have been clearly identified, and these steps and elements do not constitute an exclusive list. The method or device may also include other steps or elements.

[0047] Although the present application makes various references to certain modules in the system according to the embodiments of the present application, any number of different modules can be used and run on the user terminal and / or server. The modules are only illustrative, and different aspects of the system and method can use different modules.

[0048] Flowcharts are used in the present application to illustrate the operations performed by the system according to the embodiments of the present application. It should be understood that the preceding or following operations are not necessarily performed accurately in order. On the contrary, various steps may be processed in reverse order or simultaneously as required. Meanwhile, other operations may also be added to these processes, or a certain step or several steps of operations may be removed from these processes.

[0049] Below, the exemplary embodiments according to the present application will be described in detail with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments of the present application, and it should be understood that the present application is not limited to the exemplary embodiments described here.

[0050] Traditional pulmonary function testing methods mainly rely on the doctor's experience and manual operation, which is not only time-consuming and laborious, but also the accuracy is affected by many factors, which makes it difficult to meet the fast and accurate medical needs of modern society. Specifically, traditional pulmonary function tests usually require professional medical personnel to operate and rely on the doctor's experience and judgment, which makes the test results susceptible to personal factors and leads to inconsistent results. Although existing pulmonary function testing equipment is already quite advanced, due to the lack of intelligent data processing and analysis methods, some subtle changes or early symptoms may not be captured in a timely and accurate manner, thus affecting the accuracy of the doctor's diagnosis. Therefore, an intelligent auxiliary system for pulmonary function testing is desired.

[0051] In recent years, with the development of artificial intelligence and machine learning technologies, data-driven methods have gradually emerged in the medical field. These technologies can extract potential human lung function characteristics and patterns through large-scale data analysis, thereby realizing automated assessment and detection of human health status and lung function abnormalities, thereby assisting doctors in diagnosing lung function and judging results.

[0052] In the technical solution of the present application, an intelligent auxiliary system for lung function testing is proposed. Figure 1 4 is a block diagram of an intelligent assistance system for lung function testing according to an embodiment of the present application. Figure 2 FIG. 1 is a data flow diagram of an intelligent auxiliary system for lung function testing according to an embodiment of the present application. Figure 1 and Figure 2 As shown, according to an embodiment of the present application, an intelligent auxiliary system 300 for lung function detection includes: a vital capacity data acquisition module 310, used to obtain a time queue of vital capacity collected by a peak flow meter and a spirometer; a vital capacity data decomposition module 320, used to perform empirical mode decomposition on the time queue of vital capacity to obtain a set of vital capacity intrinsic mode functions and a set of expiratory flow rate intrinsic mode functions; a vital capacity intrinsic mode frequency domain semantic feature extraction module 330, used to perform feature extraction on the set of vital capacity intrinsic mode functions to obtain a set of vital capacity intrinsic mode frequency domain semantic coding features; a vital capacity decomposition kernel feature dynamic aggregation module 340, used to perform kernel feature dynamic aggregation analysis on the set of vital capacity intrinsic mode frequency domain semantic coding features to obtain vital capacity time series kernel features; a lung function detection module 350, used to perform lung function detection based on the vital capacity time series kernel features to determine the detection results.

[0053] In particular, the vital capacity data collection module 310 is used to obtain the time sequence of the vital capacity collected by the peak flow meter and the spirometer.

[0054] In particular, the vital capacity data decomposition module 320 is used to perform empirical mode decomposition on the time queue of the vital capacity to obtain a set of vital capacity intrinsic mode functions and a set of expiratory flow rate intrinsic mode functions. It should be understood that empirical mode decomposition is an adaptive signal processing method, which can decompose complex nonlinear and non-stationary signals into several intrinsic mode functions. Each intrinsic mode function represents a specific frequency component of the original vital capacity time series information, and these intrinsic mode functions can better reflect the local characteristics and frequency domain information of the vital capacity time series data. By performing empirical mode decomposition processing on the time queue data of vital capacity, more detailed lung function performance characteristics can be obtained, providing a rich information basis for subsequent deep learning models and lung function detection tasks.

[0055] In particular, the vital capacity intrinsic modal frequency domain semantic feature extraction module 330 is used to extract features from the set of vital capacity intrinsic modal functions to obtain a set of vital capacity intrinsic modal frequency domain semantic coding features. In a specific example of the present application, the set of vital capacity intrinsic modal functions is input into an intrinsic modal feature extractor based on a 1D-CNN model to obtain a set of vital capacity intrinsic modal frequency domain semantic coding feature vectors as the set of vital capacity intrinsic modal frequency domain semantic coding features. Through the processing of the intrinsic modal feature extractor based on the 1D-CNN model, the frequency domain implicit semantic coding features in each vital capacity intrinsic modal function can be captured respectively, which helps to use the local frequency domain semantic features in the vital capacity time series data to identify subtle changes and abnormal patterns of lung function, such as early signs of diseases such as asthma and chronic obstructive pulmonary disease (COPD), thereby assisting in lung function status detection and user health management.

[0056] In particular, the vital capacity decomposition kernel feature dynamic aggregation module 340 is used to perform kernel feature dynamic aggregation analysis on the set of vital capacity intrinsic modal frequency domain semantic coding features to obtain vital capacity time series kernel features. It should be understood that since the individual vital capacity intrinsic modal frequency domain semantic coding features in the set of vital capacity intrinsic modal frequency domain semantic coding features have an association relationship and interactive information based on the overall vital capacity time series semantics, and in the process of pulmonary function detection, these different vital capacity intrinsic modal frequency domain semantics have different importance and contribution to the subsequent feature aggregation representation and detection tasks. However, traditional feature aggregation methods are usually difficult to directly measure the specific contributions of different vital capacity intrinsic modal frequency domain semantic features to subsequent tasks, and it is also difficult to capture the complex internal structure and nonlinear relationship between different vital capacity intrinsic modal frequency domain semantic features, and simple linear transformation is not enough to fully describe these frequency domain semantic association feature information. Based on this, in order to extract a compact and information-rich kernel feature aggregation representation that can more accurately reflect the essential characteristics of the lung function state from the set of vital capacity intrinsic modal frequency domain semantic coding features, so as to better perform subsequent lung function detection tasks, in the technical solution of the present application, the set of the vital capacity intrinsic modal frequency domain semantic coding features is further subjected to kernel feature dynamic aggregation analysis to obtain the vital capacity time series kernel features. In particular, the process of kernel feature dynamic aggregation analysis can extract a compact and information-rich kernel feature that can more accurately reflect the essential characteristics of the data from the set of the vital capacity intrinsic modal frequency domain semantic coding features. In the process of kernel feature dynamic aggregation analysis, the feature force field can be used as an abstract space, in which each vital capacity intrinsic modal frequency domain semantic coding feature has a certain energy or influence, and this influence includes both static attributes (i.e., the information carried by the vital capacity intrinsic modal frequency domain semantic coding feature itself) and dynamic attributes (i.e., the interaction between each vital capacity intrinsic modal frequency domain semantic coding feature). Dynamic analysis refers to the process of modeling the interaction patterns between these vital capacity intrinsic modal frequency domain semantic encoding features, while kernel feature aggregation refers to converting the original set of vital capacity intrinsic modal frequency domain semantic encoding features into a highly generalized core representation through a series of computational operations. In a specific example of the present application, Figure 3As shown, the vital capacity decomposition kernel feature dynamic aggregation module 340 includes: an intrinsic modal frequency domain semantic kernel factor calculation unit 341, which is used to calculate the static energy potential factor and dynamic energy potential factor of each vital capacity intrinsic modal frequency domain semantic coding feature in the set of vital capacity intrinsic modal frequency domain semantic coding features to obtain a set of vital capacity intrinsic modal frequency domain semantic static energy potential factors and a set of vital capacity intrinsic modal frequency domain semantic dynamic energy potential factors; a vital capacity time series kernel feature aggregation unit 342, which is used to perform kernel semantic aggregation on the set of vital capacity intrinsic modal frequency domain semantic coding features based on the set of vital capacity intrinsic modal frequency domain semantic static energy potential factors and the set of vital capacity intrinsic modal frequency domain semantic dynamic energy potential factors to obtain the vital capacity time series kernel feature.

[0057] Specifically, the intrinsic modal frequency domain semantic kernel factor calculation unit 341 is used to calculate the static potential factor and dynamic potential factor of each vital capacity intrinsic modal frequency domain semantic coding feature in the set of vital capacity intrinsic modal frequency domain semantic coding features to obtain a set of vital capacity intrinsic modal frequency domain semantic static potential factors and a set of vital capacity intrinsic modal frequency domain semantic dynamic potential factors. In the technical solution of the present application, first, the positional mean vector of the set of vital capacity intrinsic modal frequency domain semantic coding feature vectors is calculated to obtain the initial aggregate representation vector of the vital capacity intrinsic modal frequency domain semantics; the positional mean vector of the set of vital capacity intrinsic modal frequency domain semantic coding feature vectors is calculated to construct a center point or typical example that reflects the overall characteristics of the entire vital capacity intrinsic modal frequency domain semantic coding feature set. This process helps to reduce noise and smooth data features. Next, the static potential factor of each vital capacity intrinsic modal frequency domain semantic coding feature vector in the set of vital capacity intrinsic modal frequency domain semantic coding feature vectors is calculated to obtain the set of static potential factors of vital capacity intrinsic modal frequency domain semantics; that is, for each individual vital capacity intrinsic modal frequency domain semantic coding feature vector, its static potential factor is calculated. This step is intended to evaluate the inherent properties or importance measure of each vital capacity intrinsic modal frequency domain semantic coding feature vector when it is independent of other vectors. The static potential factor is used to distinguish which vital capacity intrinsic modal frequency domain semantic coding features themselves carry more information. Furthermore, the dynamic potential factor of each vital capacity intrinsic modal frequency domain semantic coding feature vector in the set of vital capacity intrinsic modal frequency domain semantic coding feature vectors is calculated relative to the vital capacity intrinsic modal frequency domain semantic initial aggregation representation vector to obtain the set of dynamic potential factors of vital capacity intrinsic modal frequency domain semantics. Here, considering the interaction between the individual vital capacity intrinsic modal frequency domain semantic encoding feature vector and the overall data set, it is necessary to calculate the dynamic potential factor of each vital capacity intrinsic modal frequency domain semantic encoding feature vector relative to the vital capacity intrinsic modal frequency domain semantic initial aggregate representation vector. The dynamic potential factor is used to help capture the relative relationship between the data, that is, the association relationship and complex interaction pattern between each vital capacity intrinsic modal frequency domain semantic feature and the initial aggregate representation. This interactive semantics contains hidden, nonlinear relationships that may be ignored in traditional feature extraction methods.

[0058] Among them, the specific steps of calculating the static energy potential factor of each vital capacity inherent modal frequency domain semantic coding feature vector in the set of vital capacity inherent modal frequency domain semantic coding feature vectors include: calculating the mean and standard deviation of the vital capacity inherent modal frequency domain semantic coding feature vector to obtain the vital capacity inherent modal frequency domain semantic mean and the vital capacity inherent modal frequency domain semantic standard deviation; calculating the positional difference between the vital capacity inherent modal frequency domain semantic coding feature vector and the vital capacity inherent modal frequency domain semantic mean, and then dividing it by the vital capacity inherent modal frequency domain semantic standard deviation by position point to obtain the vital capacity inherent modal frequency domain semantic distribution representation vector; calculating the fourth power of each positional feature value in the vital capacity inherent modal frequency domain semantic distribution representation vector, and then calculating the mean of the obtained fourth power modulation vector to obtain the vital capacity inherent modal frequency domain semantic static energy potential factor.

[0059] More specifically, the specific steps of calculating the dynamic energy potential factor of each vital capacity intrinsic modal frequency domain semantic coding feature vector in the set of the vital capacity intrinsic modal frequency domain semantic coding feature vector relative to the vital capacity intrinsic modal frequency domain semantic initial aggregation representation vector include: calculating the multiplication between the vital capacity intrinsic modal frequency domain semantic initial aggregation representation vector and the vital capacity intrinsic modal frequency domain semantic initial clustering weight matrix to obtain the vital capacity intrinsic modal frequency domain semantic initial aggregation representation modulation vector; calculating the vital capacity intrinsic modal frequency domain semantic coding feature vector and the vital capacity intrinsic modal frequency domain semantic initial aggregation representation modulation vector. Multiplication between modal frequency domain semantic weight matrices to obtain the vital capacity inherent modal frequency domain semantic modulation vector; point-wise multiplication of the vital capacity inherent modal frequency domain semantic initial aggregation representation modulation vector and the vital capacity inherent modal frequency domain semantic modulation vector, and then point-wise multiplication with the vital capacity inherent modal frequency domain semantic bias vector to obtain the vital capacity inherent modal frequency domain semantic dynamic modulation representation vector; calculate the square of the Frobenius norm of the vital capacity inherent modal frequency domain semantic dynamic modulation representation vector to obtain the vital capacity inherent modal frequency domain semantic dynamic potential factor.

[0060] Specifically, the vital capacity time series kernel feature aggregation unit 342 is used to perform kernel semantic aggregation on the set of vital capacity inherent modal frequency domain semantic coding features based on the set of vital capacity inherent modal frequency domain semantic static energy potential factors and the set of vital capacity inherent modal frequency domain semantic dynamic energy potential factors to obtain the vital capacity time series kernel feature. In the technical solution of the present application, firstly, based on the static energy potential factor of the vital capacity intrinsic modal frequency domain semantics of each vital capacity intrinsic modal frequency domain semantic coding feature vector in the set of the vital capacity intrinsic modal frequency domain semantic coding feature vector, and the dynamic energy potential factor of the vital capacity intrinsic modal frequency domain semantic coding feature vector, the superposition state energy potential factor of each vital capacity intrinsic modal frequency domain semantic coding feature vector in the set of the vital capacity intrinsic modal frequency domain semantic coding feature vector is determined to obtain a set of superposition state energy potential factors of the vital capacity intrinsic modal frequency domain semantics; that is, after obtaining the static and dynamic energy potential factors corresponding to each vital capacity intrinsic modal frequency domain semantic coding feature vector, the next step is to combine these two types of information to form a superposition state energy potential factor. This comprehensive index reflects the compound influence of a single vital capacity intrinsic modal frequency domain semantic feature in the entire system. Next, the set of the vital capacity intrinsic modal frequency domain semantic superposition state potential factors is input into the potential factor gated weighting network to obtain the set of vital capacity intrinsic modal frequency domain semantic feature superposition potential weight coefficients; here, these superposition state potential factors are input into the potential factor gated weighting network, which can automatically learn the best weight allocation rule according to the data. That is, in order to enhance the interpretability and transparency of the model, the attention mechanism is introduced at this stage to explicitly mark the vital capacity intrinsic modal frequency domain semantic features or their combinations that have the greatest impact on the final results and subsequent lung function detection tasks. Finally, based on the set of the vital capacity intrinsic modal frequency domain semantic feature superposition potential weight coefficients, the position-weighted sum of the set of the vital capacity intrinsic modal frequency domain semantic encoding feature vectors is calculated to obtain the vital capacity time series kernel feature vector as the vital capacity time series kernel feature. By performing position-weighted summation on the set of the vital capacity intrinsic modal frequency domain semantic encoding feature vectors, the final vital capacity time series kernel feature vector can be obtained as the vital capacity time series kernel feature. This is an effective dimensionality reduction method that removes redundancy while retaining the main semantic features of the vital capacity intrinsic modal frequency domain. In addition, such processing can better capture the subtle differences and potential connections between the semantic features of the vital capacity intrinsic modal frequency domain, providing a basis for subsequent auxiliary lung function testing.

[0061] Among them, based on the vital capacity inherent modal frequency domain semantic static energy potential factor and the vital capacity inherent modal frequency domain semantic dynamic energy potential factor of each vital capacity inherent modal frequency domain semantic coding feature vector in the set of vital capacity inherent modal frequency domain semantic coding feature vectors, the specific steps of determining the superposition state energy potential factor of each vital capacity inherent modal frequency domain semantic coding feature vector in the set of vital capacity inherent modal frequency domain semantic coding feature vectors include: calculating the vital capacity inherent modal frequency domain semantic static energy potential factor of the vital capacity inherent modal frequency domain semantic coding feature vector with the natural constant e as the base and The natural exponential function value of the exponential power is used to obtain the static energy potential class support factor of the vital capacity inherent modal frequency domain semantics; the natural exponential function value of the exponential power is used to obtain the dynamic energy potential class support factor of the vital capacity inherent modal frequency domain semantics of the vital capacity inherent modal frequency domain semantic coding feature vector, with the natural constant e as the base and the square of the dynamic energy potential class support factor of the vital capacity inherent modal frequency domain semantics. The sum of the static energy potential class support factor of the vital capacity inherent modal frequency domain semantics and the dynamic energy potential class support factor of the vital capacity inherent modal frequency domain semantics is calculated to obtain the superposition state energy potential factor of the vital capacity inherent modal frequency domain semantics.

[0062] More specifically, the specific steps of inputting the set of vital capacity intrinsic modal frequency domain semantic superposition state energy potential factors into the energy potential factor gated weighting network include: calculating the natural exponential function value with the natural constant e as the base and the negative number of the vital capacity intrinsic modal frequency domain semantic superposition state energy potential factor as the exponential power to obtain the vital capacity intrinsic modal frequency domain semantic superposition state energy potential class support factor; calculating the inverse of the sum of the vital capacity intrinsic modal frequency domain semantic superposition state energy potential class support factor and the constant one to obtain the vital capacity intrinsic modal frequency domain semantic feature superposition energy potential normalized representation coefficient; masking the vital capacity intrinsic modal frequency domain semantic feature superposition energy potential normalized representation coefficient to obtain the vital capacity intrinsic modal frequency domain semantic feature superposition energy potential weight coefficient. Among them, the process of masking the normalized representation coefficient of the superposition energy potential of the inherent modal frequency domain semantic feature of vital capacity includes: in response to the normalized representation coefficient of the superposition energy potential of the inherent modal frequency domain semantic feature of vital capacity being greater than a predetermined threshold, taking the normalized representation coefficient of the superposition energy potential of the inherent modal frequency domain semantic feature of vital capacity as the original value, otherwise setting it to 0, so as to obtain the weight coefficient of the superposition energy potential of the inherent modal frequency domain semantic feature of vital capacity.

[0063] In summary, the kernel feature dynamic aggregation analysis is performed on the set of the vital capacity intrinsic modal frequency domain semantic coding features to obtain the vital capacity time series kernel feature, including: performing the kernel feature dynamic aggregation analysis on the set of the vital capacity intrinsic modal frequency domain semantic coding feature vectors using the following feature dynamic aggregation formula to obtain the vital capacity time series kernel feature;

[0064] The dynamic aggregation formula of the feature is:

[0065] ;

[0066] ;

[0067] ;

[0068] ;

[0069] ;

[0070] ;

[0071] ;

[0072] ;

[0073] ;

[0074] in, is a set of semantically encoded feature vectors of the vital capacity intrinsic modal frequency domain, are the first, second, ..., and third vectors in the set of the intrinsic modal frequency domain semantic encoding feature vectors of vital capacity. ,..., The frequency domain semantic encoding feature vector of vital capacity intrinsic modality, is the number of vectors in the set of the intrinsic modal frequency domain semantic coding feature vectors of vital capacity, is the initial aggregate representation vector of the intrinsic modal frequency domain semantics of vital capacity, It is The eigenvalues ​​of each position in the semantic encoding feature vector of the intrinsic modal frequency domain of vital capacity, and They are The mean and standard deviation of yes The number of eigenvalues ​​in , yes The corresponding vital capacity intrinsic modal frequency domain semantic static energy potential factor, , and are the initial clustering weight matrix of the vital capacity intrinsic modal frequency domain semantics, the weight matrix of the vital capacity intrinsic modal frequency domain semantics feature, and the vital capacity intrinsic modal frequency domain semantics bias vector, respectively. It means point multiplication by position. represents the square of the Frobenius norm of a vector, yes The corresponding vital capacity intrinsic modal frequency domain semantic dynamic energy potential factor, yes The corresponding vital capacity intrinsic modal frequency domain semantic superposition state energy potential factor, yes The corresponding vital capacity intrinsic modal frequency domain semantic feature superposition energy potential normalized representation coefficient, Indicates masking processing. is the predetermined threshold, yes The corresponding vital capacity intrinsic modal frequency domain semantic feature superposition energy potential weight coefficient, is the vital capacity time series kernel feature vector.

[0075] In particular, the pulmonary function detection module 350 is used to perform pulmonary function detection based on the vital capacity time series kernel feature to determine the test result. In a specific example of the present application, the vital capacity time series kernel feature vector is input into a classifier-based pulmonary function detection module to obtain a test result, and the test result is used to indicate whether there is an abnormality. That is, the classification process is performed using the intrinsic modal frequency domain semantic time series kernel aggregation representation of the vital capacity to automatically perform the user's pulmonary function test and give a test result of whether there is an abnormality in the pulmonary function. This helps to identify abnormal patterns of the user's pulmonary function, such as early signs of diseases such as asthma and chronic obstructive pulmonary disease (COPD). In this way, it is possible to achieve automated and intelligent detection of the user's pulmonary function status, thereby assisting the user himself or medical personnel so that they can promptly discover early or hidden pulmonary function abnormalities, providing strong support for clinical diagnosis and personal health management.

[0076] Preferably, inputting the vital capacity time series kernel feature vector into a classifier-based lung function detection module to obtain a detection result comprises:

[0077] Determine the number of zero eigenvalues ​​in the vital capacity time series kernel eigenvector, and respectively calculate the reciprocal of the logarithm of the number of zero eigenvalues ​​with base two and the exponential value of the reciprocal of the number of zero eigenvalues ​​with base natural constant to obtain a first vital capacity time series kernel representation value and a second vital capacity time series kernel representation value;

[0078] Calculate a power function of each eigenvalue of the vital capacity time series kernel eigenvector with the inverse of the number of zero eigenvalues ​​as an exponent to obtain a vital capacity time series kernel structure state vector;

[0079] Performing a dot multiplication of the vital capacity timing kernel structure state vector and the first vital capacity timing kernel representation value to obtain a vital capacity timing kernel information representation vector;

[0080] Performing dot multiplication of the autocorrelation matrix of the vital capacity time series kernel structure state vector with the second vital capacity time series kernel representation value and weight hyperparameters to obtain a vital capacity time series kernel regression understanding matrix;

[0081] Performing matrix multiplication of the vital capacity time series kernel information representation vector and the vital capacity time series kernel regression understanding matrix to obtain an optimized vital capacity time series kernel feature vector;

[0082] The optimized vital capacity time series kernel feature vector is input into a classifier-based lung function detection module to obtain a detection result.

[0083] That is, the first vital capacity time series kernel representation value and the second vital capacity time series kernel representation value are expressed as:

[0084] ;

[0085] ;

[0086] in is the number of zero eigenvalues ​​in the kernel eigenvector of the vital capacity time series, is the first vital capacity time series kernel representation value, is the second vital capacity time series kernel representation value, It is an exponential function with a natural constant as base;

[0087] And the optimized vital capacity time series kernel feature vector is expressed as:

[0088] ;

[0089] ;

[0090] ;

[0091] in, is the vital capacity time series kernel feature vector, is the weight hyperparameter, is the vital capacity time series kernel regression comprehension matrix, is the vital capacity time series kernel information representation vector, is the number of zero eigenvalues ​​in the kernel eigenvector of the vital capacity time series, is the first vital capacity time series kernel representation value, is the second vital capacity time series kernel representation value, represents a power function in which each eigenvalue of the vital capacity time series kernel eigenvector is calculated with the inverse of the number of zero eigenvalues ​​as the exponent, represents the transpose of a vector, It is the point multiplication by position. is matrix multiplication, is the optimized vital capacity time series kernel feature vector.

[0092] Here, when each vital capacity intrinsic modal frequency domain semantic coding feature vector in the set of vital capacity intrinsic modal frequency domain semantic coding feature vectors respectively represents the frequency domain associated semantic features of the vital capacity intrinsic modal function in the local modal function domain, when performing kernel feature extraction based on feature force field dynamic analysis, the full modal function domain kernel feature representation of the vital capacity time series kernel feature vector will also have a long-range class regression understanding representation imbalance relative to the feature distribution field selective state space of the local modal function domain, thereby deviating from the source function modal semantic distribution representation class understanding and affecting the accuracy of the classification results.

[0093] Based on this, the present application realizes effective long-range modeling of the vital capacity time series kernel feature vector under field linear complexity by modeling the vector field zero dimension of the vital capacity time series kernel feature vector as a global field redundant dependency, so as to perform structured correlation self-distillation between information representation and class understanding in a selective state space based on feature distribution in the feature regression process of the vital capacity time series kernel feature vector, thereby maintaining the overall information complexity of the feature distribution of the vital capacity time series kernel feature vector while promoting the long-range dynamic class understanding balance of its feature convergence, so as to improve the class understanding consistency between the classification regression and feature extraction processes, and improve the accuracy of the test results obtained by the vital capacity time series kernel feature vector input based on the classifier lung function detection module. In this way, it is possible to realize the automatic and intelligent detection of the user's lung function status, so as to assist the user himself or medical personnel so that they can timely discover early or hidden lung function abnormalities, and provide strong support for clinical diagnosis and personal health management.

[0094] As described above, the intelligent assistance system 300 for pulmonary function detection according to the embodiment of the present application can be implemented in various wireless terminals, such as a server with an intelligent assistance algorithm for pulmonary function detection. In one possible implementation, the intelligent assistance system 300 for pulmonary function detection according to the embodiment of the present application can be integrated into the wireless terminal as a software module and / or a hardware module. For example, the intelligent assistance system 300 for pulmonary function detection can be a software module in the operating system of the wireless terminal, or it can be an application developed for the wireless terminal; of course, the intelligent assistance system 300 for pulmonary function detection can also be one of the many hardware modules of the wireless terminal.

[0095] Alternatively, in another example, the intelligent assistance system 300 for pulmonary function testing and the wireless terminal may also be separate devices, and the intelligent assistance system 300 for pulmonary function testing may be connected to the wireless terminal via a wired and / or wireless network and transmit interactive information in accordance with an agreed data format.

[0096] The embodiments of the present disclosure have been described above, and the above description is exemplary, not exhaustive, and is not limited to the disclosed embodiments. Many modifications and changes will be apparent to those of ordinary skill in the art without departing from the scope and spirit of the described embodiments. The selection of terms used herein is intended to best explain the principles of the embodiments, practical applications, or improvements to the technology in the market, or to enable other persons of ordinary skill in the art to understand the embodiments disclosed herein.

Claims

1. An intelligent auxiliary system for lung function testing, characterized in that: include: A vital capacity data acquisition module, used for acquiring a time queue of vital capacity acquired by a peak flow meter and a spirometer; A vital capacity data decomposition module, used for performing empirical mode decomposition on the time queue of the vital capacity to obtain a set of vital capacity intrinsic mode functions and a set of expiratory flow rate intrinsic mode functions; A vital capacity intrinsic modal frequency domain semantic feature extraction module, used for extracting features from the set of vital capacity intrinsic modal functions to obtain a set of vital capacity intrinsic modal frequency domain semantic coding features; A vital capacity decomposition kernel feature dynamic aggregation module is used to perform kernel feature dynamic aggregation analysis on the set of vital capacity intrinsic modal frequency domain semantic coding features to obtain vital capacity time series kernel features; wherein, the vital capacity decomposition kernel feature dynamic aggregation module includes: an intrinsic modal frequency domain semantic kernel factor calculation unit, which is used to calculate the static energy potential factor and dynamic energy potential factor of each vital capacity intrinsic modal frequency domain semantic coding feature in the set of vital capacity intrinsic modal frequency domain semantic coding features to obtain a set of vital capacity intrinsic modal frequency domain semantic static energy potential factors and a set of vital capacity intrinsic modal frequency domain semantic dynamic energy potential factors; a vital capacity time series kernel feature aggregation unit, which is used to perform kernel semantic aggregation on the set of vital capacity intrinsic modal frequency domain semantic coding features based on the set of vital capacity intrinsic modal frequency domain semantic static energy potential factors and the set of vital capacity intrinsic modal frequency domain semantic dynamic energy potential factors to obtain the vital capacity time series kernel features; The pulmonary function detection module is used to perform pulmonary function detection based on the vital capacity time series kernel feature to determine the detection result.

2. The intelligent auxiliary system for lung function testing according to claim 1, characterized in that: The vital capacity intrinsic modal frequency domain semantic feature extraction module is used to: input the set of vital capacity intrinsic modal functions into the intrinsic modal feature extractor based on the 1D-CNN model to obtain a set of vital capacity intrinsic modal frequency domain semantic coding feature vectors as the set of vital capacity intrinsic modal frequency domain semantic coding features.

3. The intelligent auxiliary system for lung function testing according to claim 2, characterized in that: The intrinsic modal frequency domain semantic kernel factor calculation unit comprises: A mean calculation subunit, used for calculating the position mean vector of the set of the vital capacity intrinsic modal frequency domain semantic encoding feature vectors to obtain the vital capacity intrinsic modal frequency domain semantic initial aggregation representation vector; A vital capacity intrinsic modal frequency domain semantic static energy potential factor calculation subunit, used to calculate the static energy potential factor of each vital capacity intrinsic modal frequency domain semantic coding feature vector in the set of vital capacity intrinsic modal frequency domain semantic coding feature vectors to obtain the set of vital capacity intrinsic modal frequency domain semantic static energy potential factors; The vital capacity intrinsic modal frequency domain semantic dynamic energy potential factor calculation subunit is used to calculate the dynamic energy potential factor of each vital capacity intrinsic modal frequency domain semantic coding feature vector in the set of vital capacity intrinsic modal frequency domain semantic coding feature vectors relative to the vital capacity intrinsic modal frequency domain semantic initial aggregation representation vector to obtain the set of vital capacity intrinsic modal frequency domain semantic dynamic energy potential factors.

4. The intelligent auxiliary system for lung function testing according to claim 3, characterized in that: The vital capacity intrinsic modal frequency domain semantic static energy potential factor calculation subunit is used for: Calculating the mean and standard deviation of the vital capacity intrinsic modal frequency domain semantic encoding feature vector to obtain the vital capacity intrinsic modal frequency domain semantic mean and the vital capacity intrinsic modal frequency domain semantic standard deviation; Calculating the position difference between the vital capacity intrinsic modal frequency domain semantic coding feature vector and the vital capacity intrinsic modal frequency domain semantic mean, and then dividing the difference by the vital capacity intrinsic modal frequency domain semantic standard deviation by position point to obtain the vital capacity intrinsic modal frequency domain semantic distribution representation vector; After calculating the fourth power of the characteristic values ​​of each position in the vital capacity intrinsic modal frequency domain semantic distribution representation vector, the mean value of the fourth power modulation vector is calculated to obtain the vital capacity intrinsic modal frequency domain semantic static energy potential factor.

5. The intelligent auxiliary system for lung function testing according to claim 4, characterized in that: The vital capacity intrinsic modal frequency domain semantic dynamic energy potential factor calculation subunit is used for: Calculate the multiplication between the vital capacity intrinsic modal frequency domain semantic initial aggregation representation vector and the vital capacity intrinsic modal frequency domain semantic initial clustering weight matrix to obtain the vital capacity intrinsic modal frequency domain semantic initial aggregation representation modulation vector; Calculating the multiplication between the vital capacity intrinsic modal frequency domain semantic coding feature vector and the vital capacity intrinsic modal frequency domain semantic weight matrix to obtain the vital capacity intrinsic modal frequency domain semantic modulation vector; The vital capacity intrinsic modal frequency domain semantic initial aggregation representation modulation vector is multiplied by the vital capacity intrinsic modal frequency domain semantic modulation vector by position point, and then multiplied by the vital capacity intrinsic modal frequency domain semantic offset vector by position point to obtain the vital capacity intrinsic modal frequency domain semantic dynamic modulation representation vector; The square of the Frobenius norm of the vital capacity intrinsic modal frequency domain semantic dynamic modulation representation vector is calculated to obtain the vital capacity intrinsic modal frequency domain semantic dynamic energy potential factor.

6. The intelligent auxiliary system for lung function testing according to claim 5, characterized in that: The vital capacity time series kernel feature aggregation unit comprises: A vital capacity intrinsic modal frequency domain semantic superposition state energy potential factor calculation subunit, used to determine the superposition state energy potential factor of each vital capacity intrinsic modal frequency domain semantic coding feature vector in the set of vital capacity intrinsic modal frequency domain semantic coding feature vectors based on the vital capacity intrinsic modal frequency domain semantic static energy potential factor and the vital capacity intrinsic modal frequency domain semantic dynamic energy potential factor of each vital capacity intrinsic modal frequency domain semantic coding feature vector in the set of vital capacity intrinsic modal frequency domain semantic coding feature vectors to obtain a set of vital capacity intrinsic modal frequency domain semantic superposition state energy potential factors; An energy potential factor gating weighting subunit is used to input the set of energy potential factors of the vital capacity intrinsic modal frequency domain semantic superposition state into an energy potential factor gating weighting network to obtain a set of vital capacity intrinsic modal frequency domain semantic feature superposition energy potential weight coefficients; The vital capacity time series kernel feature calculation subunit is used to calculate the position-weighted sum of the set of vital capacity inherent modal frequency domain semantic coding feature vectors based on the set of potential weight coefficients superimposed on the vital capacity inherent modal frequency domain semantic features to obtain the vital capacity time series kernel feature vector as the vital capacity time series kernel feature.

7. The intelligent auxiliary system for lung function testing according to claim 6, characterized in that: The vital capacity intrinsic modal frequency domain semantic superposition state energy potential factor calculation subunit is used for: Calculate the natural exponential function value with the natural constant e as the base and the vital capacity intrinsic modal frequency domain semantic static energy potential factor of the vital capacity intrinsic modal frequency domain semantic encoding feature vector as the exponential power to obtain the vital capacity intrinsic modal frequency domain semantic static energy potential class support factor; Calculate the natural exponential function value with the natural constant e as the base and the square of the vital capacity intrinsic modal frequency domain semantic dynamic energy potential factor of the vital capacity intrinsic modal frequency domain semantic encoding feature vector as the exponential power to obtain the vital capacity intrinsic modal frequency domain semantic dynamic energy potential class support factor; The sum of the vital capacity intrinsic modal frequency domain semantic static energy potential class support factor and the vital capacity intrinsic modal frequency domain semantic dynamic energy potential class support factor is calculated to obtain the vital capacity intrinsic modal frequency domain semantic superposition state energy potential factor.

8. The intelligent auxiliary system for lung function testing according to claim 7, characterized in that: The energy potential factor gated weighting subunit comprises: The secondary subunit for calculating the vital capacity intrinsic modal frequency domain semantic superposition state energy potential class support factor is used to calculate the natural exponential function value with the natural constant e as the base and the negative number of the vital capacity intrinsic modal frequency domain semantic superposition state energy potential factor as the exponential power to obtain the vital capacity intrinsic modal frequency domain semantic superposition state energy potential class support factor; A secondary subunit for calculating the normalized representation coefficient of the superposition energy potential of the intrinsic modal frequency domain semantic features of vital capacity, which is used to calculate the inverse of the sum of the intrinsic modal frequency domain semantic superposition state energy potential class support factor of the vital capacity and a constant one to obtain the normalized representation coefficient of the superposition energy potential of the intrinsic modal frequency domain semantic features of vital capacity; A masked secondary subunit is used to perform masking processing on the normalized representation coefficient of the vital capacity intrinsic modal frequency domain semantic feature superposition energy potential to obtain a vital capacity intrinsic modal frequency domain semantic feature superposition energy potential weight coefficient; Among them, the masked secondary sub-unit is further used to: in response to the normalized representation coefficient of the superposition potential of the inherent modal frequency domain semantic features of the vital capacity being greater than a predetermined threshold, take the normalized representation coefficient of the superposition potential of the inherent modal frequency domain semantic features of the vital capacity as the original value, otherwise set it to 0, so as to obtain the superposition potential weight coefficient of the inherent modal frequency domain semantic features of the vital capacity.

9. The intelligent auxiliary system for lung function testing according to claim 8, characterized in that: The pulmonary function detection module is used to: input the vital capacity time series kernel feature vector into the classifier-based pulmonary function detection module to obtain a detection result, and the detection result is used to indicate whether there is an abnormality.

Images