Nitric oxide data analysis method and system for bronchial asthma assessment

Abstract

The invention relates to a nitric oxide data analysis method and system for bronchial asthma assessment. According to the method, exhaled air nitric oxide and pulmonary alveolar nitric oxide of a patient, omics data and clinical information are acquired in a standardized manner, and multi-dimensional features are constructed after data cleaning and fusion; comprise an inflammation level index, an inflammation region entropy, an inflammation synergy index, a glucocorticoid response factor, an infection-smoking synergy influence factor and a heredity-symptom distribution index. And outputting initial risk assessment based on an explainable elevator EBM model, and finally generating an asthma risk probability through a weighted integration formula to realize three-level layering: low / medium / high risk. The system dynamically associates treatment decisions, for example, the drug dosage and the monitoring frequency are improved when the risk is upgraded, and the medication scheme is optimized when the risk is degraded. According to the scheme, the limitation of traditional single marker static analysis is broken through, multiple mechanisms of dissection, immunity and pharmacology are fused, the evaluation precision and clinical applicability are remarkably improved, the acute attack rate is reduced, and accurate typing treatment is guided.

Description

Technical Field

[0001] The present invention relates to the field of medical diagnosis technology, and in particular to a nitric oxide data analysis method and system for bronchial asthma assessment. Background Art

[0002] There are significant limitations in the accurate assessment of bronchial asthma: existing methods analyze exhaled nitric oxide in isolation, ignoring the localization value of alveolar nitric oxide for small airway inflammation, resulting in a high rate of missed diagnosis of small airway lesions; traditional models rely on single clinical symptoms or omics data and are unable to quantify the interactive effects of smoking history, recent infection, and genetic background; mainstream tools use static thresholds to adjust treatments, which makes it difficult to respond to dynamic changes in inflammation, resulting in some patients receiving ineffective treatment. Although emerging technologies introduce gene expression to improve accuracy, genetic testing is expensive and invasive, and the black-box nature of predictive models hinders the tracing of clinical decisions, and treatment plans lack dynamic associations with environmental triggering events. There is an urgent need to develop an evaluation system that integrates spatial analysis of gas markers, dynamic environmental responses, and explainable decisions to achieve a fundamental upgrade of the asthma management paradigm. Summary of the Invention

[0003] In view of the deficiencies of the existing technology, the present invention provides a nitric oxide data analysis method and system for bronchial asthma assessment.

[0004] To achieve the above object, the present invention adopts the following technical solutions: The method for analyzing nitric oxide data for evaluating bronchial asthma comprises the following steps: S1. Collect patients' exhaled nitric oxide, alveolar nitric oxide, omics data and clinical information; S2. Data cleaning, normalization, and missing value processing are performed, and the processed exhaled nitric oxide and alveolar nitric oxide, omics data, and clinical information are integrated; S3. Construct the following features: Inflammation level index, inflammation area entropy, inflammation synergy index, glucocorticoid response factor, infection-smoking synergistic influence factor, genetic-symptom distribution index; S4, divide the training set and test set; S5, using the interpretable boosting machine EBM as the classifier and the decision tree as the base learner to iteratively train the model; S6. Use the trained EBM model to output the initial assessment value of bronchial asthma; S7. Construct a final bronchial asthma assessment formula, substitute the initial bronchial asthma assessment value and various characteristics into the final bronchial asthma assessment formula, and obtain the bronchial asthma assessment grade.

[0005] In one implementation method of the present invention, S1 includes the following specific contents: S1. Data Collection: Exhaled nitric oxide and alveolar nitric oxide data will be collected from patients using standardized testing methods and equipment to ensure data accuracy and reliability. Other omics data will also be collected, including routine blood test data and proteomics data. These data can be obtained using appropriate testing techniques and equipment. Clinical information about the patient, such as age, gender, smoking history, allergy history, family history of asthma, symptoms, lung function indicators, use of inhaled steroids, and recent respiratory infections, will be collected as auxiliary data for analysis. In one implementation method of the present invention, the data preprocessing in step S2 includes the following specific contents: S21. Data preprocessing: Preprocess the collected multi-omics data, including data cleaning, normalization, and missing value processing, to remove noise and outliers in the data and ensure data quality and consistency; S22. Data fusion: Fuse the preprocessed exhaled nitric oxide with alveolar nitric oxide data, other omics data, and clinical information to construct a multi-omics dataset. Different data fusion methods can be used, such as feature-based splicing and model-based fusion, to integrate data from different sources. The fused data can then be further analyzed and mined to extract key features and reduce the dimensionality and complexity of the data. In one implementation method of the present invention, S3 includes the following specific contents: S31. In order to accurately quantify the spatial distribution characteristics of bronchial asthma, the nitric oxide data in the dataset, exhaled nitric oxide and alveolar nitric oxide, are combined to construct the following hierarchical features: inflammation level index, inflammation level index The derivation steps are as follows: standardize the exhaled nitric oxide by the reference exhaled nitric oxide concentration of healthy people, and perform power law transformation on the standardized exhaled nitric oxide to obtain the exhaled nitric oxide weight adjustment term Then, the alveolar nitric oxide was corrected by the eosinophil count, and the corrected alveolar nitric oxide was standardized to calculate the alveolar nitric oxide-eosinophil logarithmic enhancement factor. The inflammation level index evaluation formula is obtained by multiplying the exhaled nitric oxide weight adjustment term and the alveolar nitric oxide-eosinophil logarithmic enhancement term. The specific expression is: ,in, is the exhaled nitric oxide concentration, is the reference value of exhaled nitric oxide for healthy people, is the alveolar nitric oxide concentration, is the baseline value of alveolar nitric oxide, is the exhaled nitric oxide weight adjustment factor, Eosinophil count, is the correction factor of eosinophils for alveolar nitric oxide; S32. To understand the distribution of inflammation areas, we construct inflammation area entropy. The specific derivation steps are as follows: Probabilistic modeling of exhaled nitric oxide and alveolar nitric oxide is obtained , calculate the single point entropy of exhaled nitric oxide and alveolar nitric oxide respectively , combining the two single point entropies into the regional entropy ,when When inflammation is highly concentrated in a single area, When the inflammation is evenly distributed.

[0006] S33. To capture the nonlinear synergistic effect of multi-source data, the following interactive features are designed: The specific derivation steps of the inflammatory synergy index are as follows: the age attenuation term after sigmoid transformation is multiplied by the Th2 basic synergy term plus the allergy- Saturated interaction terms , ,in is the age-dependent attenuation factor, is the weight coefficient of allergy history, is a binary variable, is the serum total immune protein concentration; The glucocorticoid response factor is derived as follows: Calculate the exhaled nitric oxide response attenuation term , multiplied by the drug dose and the inflammatory change rate to obtain the glucocorticoid response factor, which is expressed as ,in, is the daily dose of inhaled corticosteroids, for The rate of change is calculated by sliding window: ; The infection-smoking synergistic influencing factor is specifically derived as follows: Smoke disrupts pulmonary vascular homeostasis through three mechanisms: Cumulative toxic effects: Polycyclic aromatic hydrocarbons in tobacco tar continuously damage bronchial microvascular endothelial cells, causing thickening and increased fragility of the vascular basement membrane. Hemodynamic stress: Carbon monoxide (COHb>5%) significantly reduces the oxygen-carrying capacity of red blood cells (arterial oxygen partial pressure decreases by 28%) through the formation of carboxyhemoglobin, triggering compensatory pulmonary artery constriction. The risk assessment formula for smoking-induced pulmonary vascular damage is: ,in, The pathological transformation of pulmonary hemorrhage to asthma exacerbation is caused by cumulative tobacco exposure: physical blockage + iron deposition is initiated. is the carboxyhemoglobin saturation, is the mean pulmonary artery pressure, It is a marker of pulmonary hypertension. The risk assessment formula for post-hemorrhagic inflammatory outbreak is: ,in The amount of bleeding, The serum iron concentration is a dynamic integration of the "smoking → bleeding → asthma" cascade chain. This chain is not a linear accumulation, but has a biphasic amplification node and a critical mutation threshold: .

[0007] Genetic-symptom distribution index, the specific derivation process is: lung function defect item Multiplying symptoms The genetic risk switch is used to determine whether to include it, and finally a logarithmic transformation is performed. ,in, Four core symptoms ( For cough, For a breath, For chest tightness, The severity is 0-3, is the weight coefficient, is the predicted value based on age and height.

[0008] In one implementation method of the present invention, the S4 includes the following specific contents: The data is divided into a training set and a test set in a certain ratio. The training set is used to train the model, and the test set is used to evaluate the performance of the model.

[0009] In one implementation method of the present invention, S5 includes the following specific contents: S51. Use EBM as the classifier model and choose decision tree as the basic learner. EBM gradually trains each feature function in an additive manner so that its contribution can be separated and interpreted. The form of EBM is ,in The bronchial asthma assessment results output by the EBM model for the evaluator’s input data are: is the characteristic function corresponding to the i-th feature, is the pairwise interaction feature function of the i-th and j-th features, the feature function, , Implemented by decision trees; S52, train the basic learner and initialize the constant model , where L is the loss function, is the actual data of the i-th patient, is a constant representing the initial prediction value, and the expression of the loss function is ,in, is the model coefficient, To regularize the parameters, iteratively train the basic learners and combine them to form a strong learner. In each iteration, the new weak learner will be trained on the residual of the current model to gradually reduce the loss of the entire model and calculate the negative gradient. ,in, is the model obtained in the previous iteration, negative gradient Indicates the direction of error, using data Training a decision tree , calculate the step size , step length Control the model update amplitude and update the model according to the calculated step size; .

[0010] S53, perform ANOVAF test on candidate feature pairs, screening The decision tree is trained by the selected features, and the interaction function is initialized to a constant zero matrix , based on the current model (including single feature item) calculate the negative gradient for iteration, t During the round iteration, the negative gradient is ,in Output of the current model, determine the optimal step size through line search ,Update interaction function: .

[0011] In one implementation method of the present invention, S6 includes the following specific contents: Use the trained evaluation model to calculate all the data of the user to be evaluated to obtain the initial evaluation value of bronchial asthma .

[0012] In one implementation method of the present invention, the step S7 includes the following specific contents: Constructing the overall bronchial asthma assessment formula: ,in For the final evaluation results, represents the sigmoid function, Represents the weight coefficient.

[0013] The present invention also provides a nitric oxide data analysis system for evaluating bronchial asthma, which includes a data acquisition module, a preprocessing module, a fusion dimension reduction module, a feature construction module, a model training module, and an evaluation output module. The modules work together to achieve the evaluation of bronchial asthma.

[0014] The nitric oxide data analysis system for bronchial asthma assessment is implemented based on the above-mentioned nitric oxide data analysis method for bronchial asthma assessment, and specifically includes: A data acquisition module is configured to obtain the patient's exhaled nitric oxide concentration, alveolar nitric oxide concentration, blood routine test data, and proteomics data through standardized testing equipment, and simultaneously collect clinical information data sets, including age, smoking history, allergy history, lung function indicators, and glucocorticoid dosage; The preprocessing module is configured to filter noise and remove outliers from the collected exhaled nitric oxide and alveolar nitric oxide and omics data, fill missing values ​​using multiple interpolation, and unify the dimensions of multi-source data through Z-score standardization; A fusion dimensionality reduction module is configured to perform feature-level splicing of preprocessed exhaled nitric oxide with alveolar nitric oxide, omics data, and clinical information; The feature construction module is configured to perform the following calculations: generate an inflammation hierarchy index based on reference values ​​of exhaled nitric oxide and eosinophil counts in healthy individuals; calculate the entropy of inflammatory regions to quantify the spatial aggregation of inflammation by modeling the probability distribution of exhaled nitric oxide concentration; generate a drug response factor by combining the daily dose of glucocorticoids with the rate of change of exhaled nitric oxide; calculate the infection-smoking synergistic influence factor by integrating the number of years of smoking, respiratory infection status, and alveolar nitric oxide constraint terms; and construct a genetic-symptom distribution index by taking the weighted product of lung function impairment and symptoms; The model training module is configured as follows: the data is divided into training and test sets in a ratio of 7:3; the interpretable boosting machine (EBM) is used as the classifier and the decision tree is used as the base learner, and the model is iteratively trained using the gradient boosting algorithm; grid search is used to optimize hyperparameters, and the Matthews correlation coefficient (MCC) is used as the evaluation metric for cross-validation; The evaluation output module is configured to input the fusion features of the patient to be evaluated into the trained EBM model and output a bronchial asthma risk score.

[0015] Compared with the prior art, the beneficial effects of the present invention are as follows: The present invention significantly improves the accuracy and clinical applicability of bronchial asthma assessment through multi-dimensional inflammation analysis technology and dynamic interaction effect modeling. Specifically: the innovative inflammation level index integrates exhaled nitric oxide standardization, eosinophil correction and logarithmic enhancement mechanism, combined with regional entropy quantification technology based on the probability distribution of alveolar nitric oxide, breaking through the limitations of the traditional single threshold method and constructing four types of medical mechanism-driven interactive features: the inflammation synergy index integrates age-dependent immune attenuation, Th2 pathway basal activity, and the saturation effect of allergic history and immunoglobulin to quantify the synergistic amplification mechanism of Th2-type inflammation; the glucocorticoid response factor combines drug dose with the dynamic change rate of inflammation to reveal the nonlinear attenuation law of drug anti-inflammatory efficacy; the infection-smoking synergistic influence factor is the product of the number of years of smoking and the small airway function constraint term, superimposed with the infection trigger and hormone compensation effect, to characterize the acute risk multiplication mechanism of environmental exposure; the genetic-symptom distribution index is based on lung function impairment, multiplied by the weighted sum of the four core symptoms and regulated by the genetic risk switch to achieve spatial coupling modeling of genetic susceptibility and clinical symptoms. At the model architecture level, the interpretable boosting machine (EBM) is used as the core, and its additive form explicitly separates the contribution of each feature: the basic feature function uses a decision tree to visualize the effect of a single feature, and the pairwise interaction function captures the synergy of cross-modal features. Through gradient boosting iteration, it provides a basis for feature-level decision-making while ensuring accuracy. The final output bronchial asthma assessment value has been clinically verified to have three values: 1) stratified diagnostic value: scores of 0-30 / 31-70 / 71-100 correspond to mild / moderate / severe asthma, respectively; 2) treatment guidance value; and 3) risk warning value. The system integrates multi-source data collection, dynamic feature calculation, and explainable AI decision-making to complete the entire process from testing to grading reporting in a short period of time, is compatible with mainstream medical equipment, and reduces annual per capita medical expenses. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 This is a schematic diagram of the overall process of the nitric oxide data analysis method for bronchial asthma assessment in this application; Figure 2 This is a diagram showing the calculation of the final evaluation results of bronchial asthma for this application; Figure 3 This is a schematic diagram of the overall framework of the nitric oxide data analysis system used for bronchial asthma evaluation in this application. DETAILED DESCRIPTION

[0017] The technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments.

[0018] Example 1

[0019] This application provides an embodiment: Figure 1 and Figure 2 As shown, The nitric oxide data analysis method for bronchial asthma evaluation comprises the following specific steps: S1. Collect patients' exhaled nitric oxide, alveolar nitric oxide, omics data and clinical information; In this embodiment, it should be specifically explained that the acquired data includes the following specific contents: Exhaled nitric oxide and alveolar nitric oxide data are collected from patients using standardized testing methods and equipment to ensure data accuracy and reliability. Other omics data, including routine blood test data and proteomics data, are also collected. These data can be obtained using appropriate testing techniques and equipment. Clinical information about the patient, such as age, gender, smoking history, allergy history, family history of asthma, symptoms, lung function indicators, inhaled steroid use, and recent respiratory infection, is collected as auxiliary data for analysis. S2. Data cleaning, normalization, and missing value processing are performed, and the processed exhaled nitric oxide and alveolar nitric oxide, omics data, and clinical information are integrated; In this embodiment, it should be specifically explained that the preprocessing of the acquired data includes the following specific contents: Data preprocessing: Preprocess the collected multi-omics data, including data cleaning, normalization, missing value processing, etc., to remove noise and outliers in the data and ensure data quality and consistency; Data fusion: Fusing preprocessed exhaled nitric oxide with alveolar nitric oxide data, other omics data, and clinical information to construct a multi-omics dataset. Different data fusion methods, such as feature-based splicing and model-based fusion, can be used to integrate data from different sources. The fused data can then be further analyzed and mined to extract key features and reduce the dimensionality and complexity of the data. S3. Construct the following features: Inflammation level index, inflammation area entropy, inflammation synergy index, glucocorticoid response factor, infection-smoking synergistic influence factor, genetic-symptom distribution index; In this embodiment, it should be specifically explained that the construction features include the following specific contents: In order to accurately quantify the spatial distribution characteristics of bronchial asthma, the nitric oxide data in the dataset, exhaled nitric oxide and alveolar nitric oxide, were combined to construct the following hierarchical features: inflammation level index, inflammation level index The derivation steps are as follows: standardize the exhaled nitric oxide by the reference exhaled nitric oxide concentration of healthy people, perform power law transformation on the standardized exhaled nitric oxide to obtain the exhaled nitric oxide and weight adjustment term Then, the alveolar nitric oxide was corrected by the eosinophil count, and the corrected alveolar nitric oxide was standardized to calculate the alveolar nitric oxide-eosinophil logarithmic enhancement factor. ,Will The weight adjustment term is multiplied by the alveolar nitric oxide-eosinophil logarithmic enhancement term to obtain the inflammation level index evaluation formula, and its specific expression is: ,in, is the exhaled nitric oxide concentration, is the reference value of exhaled nitric oxide for healthy people, For exhaled nitric oxide, is the baseline value of alveolar nitric oxide, is the exhaled nitric oxide weight adjustment factor, preferably 0.35, Eosinophil count, The correction factor for eosinophils to alveolar nitric oxide is optimized to be 0.15. This index innovatively integrates airway anatomy and inflammatory activity non-invasively, achieving precise spatial stratification of airway inflammation through the synergistic weighting of exhaled nitric oxide and alveolar nitric oxide. In order to understand the distribution of inflammation areas, the inflammation area entropy is constructed. The specific derivation steps are as follows: Probabilistic modeling of exhaled nitric oxide and alveolar nitric oxide is obtained , calculate the single point entropy of exhaled nitric oxide and alveolar nitric oxide respectively , combining the two single point entropies into the regional entropy ,when When inflammation is highly concentrated in a single area, When the inflammation is evenly distributed, the innovative value of this indicator lies in the information-theoretic quantification of spatial inflammatory heterogeneity, which solves the problem that traditional technologies only focus on gene expression and ignore spatial distribution. The logarithmic transformation compresses the complex three-dimensional distribution into an operational indicator of [0,1], allowing doctors to intuitively identify the inflammation diffusion pattern.

[0020] To capture the nonlinear synergistic effects of multi-source data, the following interactive features are designed: The specific derivation steps of the inflammatory synergy index are as follows: the age attenuation term after sigmoid transformation is multiplied by the Th2 basic synergy term plus the allergy- Saturated interaction terms , ,in is the age-dependent attenuation factor, is the weight coefficient of allergy history, preferably 0.28, is a binary variable, The advantage of this formula is that it analyzes the multi-level coordination of the Th2 pathway, integrates the core biomarkers of eosinophilic inflammation through the FeNO×EOS product term, and accurately captures the difference in Th2 activity between adolescent asthma (β<0) and adult asthma (β>0) using the age-dependent sigmoid attenuation factor. The specific derivation process of glucocorticoid response factor is as follows: Calculate Response attenuation term , multiplied by the drug dose and the inflammatory change rate to obtain the glucocorticoid response factor, which is expressed as ,in, is the daily dose of inhaled corticosteroids, for The rate of change is calculated by sliding window: The core advantage of this formula lies in the dynamic quantification of anti-inflammatory treatment efficiency. By introducing the time derivative of the inflammation level index, it captures the rate of inflammation resolution under hormone therapy in real time, addressing the deficiency of traditional static models in reflecting the timeliness of treatment. The design of the nonlinear inhibition term is particularly critical: when FeNO is greater than 30ppb, the output value increases sharply from 0.5 to 0.9, accurately simulating the "hormone resistance in high inflammation state" phenomenon observed clinically. The product structure with the ICS dose further achieves dual regulation, encouraging sufficient medication to suppress inflammation while warning of the dose-effectiveness decline under high inflammation. The specific derivation process of the infection-smoking synergistic influencing factor is as follows: Smoke destroys pulmonary vascular homeostasis through three mechanisms: tobacco smoke first directly damages bronchial microvascular endothelial cells through cumulative toxic effects. Among them, the polycyclic aromatic hydrocarbons contained in tobacco tar can sustainably induce mitochondrial DNA breakage in endothelial cells, causing abnormal thickening of the vascular basement membrane by 3 to 5 times and significantly increasing its mechanical fragility; at the same time, hemodynamic stress serves as the second mechanism. When the carboxyhemoglobin saturation COHb exceeds 5%, the arterial oxygen partial pressure can drop by up to 28%, thereby triggering a compensatory pulmonary artery contraction response; and the synergistic effect of infection factors and smoking constitutes the third destructive mechanism, which is typically manifested by the increased replication efficiency of influenza virus in an environment of high nicotine-induced TLR4 expression. The risk assessment formula for smoking-induced pulmonary vascular damage is: ,in, The pathological transformation of pulmonary hemorrhage to asthma exacerbation is caused by cumulative tobacco exposure: physical blockage + iron deposition is initiated. is the carboxyhemoglobin saturation, is the mean pulmonary artery pressure, It is a marker of pulmonary hypertension. The risk assessment formula for post-hemorrhagic inflammatory outbreak is: ,in The amount of bleeding, The serum iron concentration is a dynamic integration of the "smoking → bleeding → asthma" cascade chain. This chain is not a linear accumulation, but has a biphasic amplification node and a critical mutation threshold: The model accurately locates the intervention window for pulmonary vascular rupture and acute asthma attacks, providing dynamic navigation for clinical stratified intervention.

[0021] Genetic-symptom distribution index, the specific derivation process is: lung function defect item Multiplying symptoms The genetic risk switch is used to determine whether to include it, and finally a logarithmic transformation is performed. ,in, Four core symptoms ( For cough, For a breath, For chest tightness, The severity is 0-3, is the weight coefficient, preferably , This formula is based on age- and height-based predictions. Its innovation lies in transforming familial genetic risk into a synergistic amplifier of symptoms and lung function. The design of family history as a genetic switch aligns with the medical consensus that genetic background is a necessary but not sufficient condition for the development of asthma, ensuring that the synergistic effect of symptoms and lung function is activated only when genetic risk is present. The symptom integral term highlights clinical value through differentiated weighting, while the introduction of a logarithmic function cleverly addresses the product explosion problem (e.g., full symptom score + 30% lung function deficit → 1.8), compressing eigenvalues ​​to a clinically interpretable range of [0, 3] while preserving the nonlinear amplification effect of genetic risk on symptom burden.

[0022] S4, divide the training set and test set; In this embodiment, it should be noted that the division of the training set and the test set includes the following specific contents: the data is divided into a training set and a test set, and the division is performed according to a certain ratio. In this embodiment, the ratio is 70%-30%. The training set is used to train the model, and the test set is used to evaluate the performance of the model.

[0023] S5, using the interpretable boosting machine EBM as the classifier and the decision tree as the base learner to iteratively train the model; In this embodiment, it should be specifically explained that the training of the interpretable model includes the following specific contents: using EBM as the classifier model, selecting the decision tree as the basic learner, and EBM gradually training each feature function in an additive manner so that its contribution can be separated and explained. The form of EBM is ,in The bronchial asthma assessment results output by the EBM model for the evaluator’s input data are: is the characteristic function corresponding to the i-th feature, is the pairwise interaction feature function of the i-th and j-th features, the feature function, , Implemented by decision trees; Train the base learner and initialize the constant model , where L is the loss function, is the actual data of the i-th patient, is a constant representing the initial prediction value, and the expression of the loss function is ,in, is the model coefficient, To regularize the parameters, iteratively train the basic learners and combine them to form a strong learner. In each iteration, the new weak learner will be trained on the residual of the current model to gradually reduce the loss of the entire model and calculate the negative gradient. ,in, is the model obtained in the previous iteration, negative gradient Indicates the direction of error, using data Training a decision tree , calculate the step size , step length Control the model update amplitude and update the model according to the calculated step size; .

[0024] Perform ANOVAF test on candidate feature pairs to screen The decision tree is trained by the selected features, and the interaction function is initialized to a constant zero matrix , based on the current model (including single feature item) calculate the negative gradient for iteration, t During the round iteration, the negative gradient is ,in Output of the current model, determine the optimal step size through line search ,Update interaction function: .

[0025] S6. Use the trained EBM model to output the initial assessment value of bronchial asthma; In this embodiment, it should be noted that S7 includes the following specific contents: Use the trained evaluation model to calculate all the data of the user to be evaluated to obtain the initial evaluation value of bronchial asthma , construct the final bronchial asthma evaluation formula, substitute the initial bronchial asthma evaluation value and various characteristics into the final bronchial asthma evaluation formula, and obtain the bronchial asthma evaluation grade.

[0026] In this embodiment, it should be noted that S7 includes the following specific contents: Constructing the overall bronchial asthma assessment formula: ,in For the final evaluation results, represents the sigmoid function, Represents the weight coefficient, preferably .

[0027] Example 2

[0028] like Figure 3 As shown, this embodiment provides a nitric oxide data analysis system for bronchial asthma assessment, which specifically includes: The data acquisition module is configured to obtain the patient's exhaled nitric oxide concentration, alveolar nitric oxide concentration, blood routine test data and proteomics data through standardized testing equipment, and simultaneously collect clinical information data sets, including age, smoking history, allergy history, lung function indicators and glucocorticoid dosage; the preprocessing module is configured to filter noise and eliminate outliers on the collected exhaled nitric oxide and alveolar nitric oxide and omics data, fill missing values ​​with multiple interpolation methods, and achieve multi-source data dimension unification through Z-score standardization; the fusion dimension reduction module is configured to perform feature-level splicing of the preprocessed exhaled nitric oxide with alveolar nitric oxide, omics data and clinical information; the feature construction module is configured to perform the following calculations: based on the reference value of exhaled nitric oxide and eosinophil count of healthy people, generate an inflammation level index By modeling the probability distribution of alveolar nitric oxide concentration, the entropy of the inflammatory area is calculated to quantify the spatial aggregation of inflammation; the daily dose of glucocorticoids is combined with the change rate of exhaled nitric oxide to generate a drug response factor; the number of years of smoking, respiratory infection status and alveolar nitric oxide constraint terms are integrated to calculate the infection-smoking synergistic influencing factor; the genetic-symptom distribution index is constructed by the weighted product of lung function defect terms and symptoms; the model training module is configured as follows: the data is divided into training set and test set in a ratio of 7:3; the interpretable boosting machine EBM is used as the classifier and the decision tree is used as the base learner, and the model is iteratively trained through the gradient boosting algorithm; the hyperparameters are optimized using grid search, and the Matthews correlation coefficient MCC is used as the evaluation indicator for cross-validation; the evaluation output module is configured to input the fusion features of the patient to be evaluated into the trained EBM model and output the bronchial asthma risk score; Example 3

[0029] This embodiment provides an electronic device, including a processor and a memory, wherein the memory contains a computer program that can be called by the processor.

[0030] The processor executes the above-mentioned nitric oxide data analysis method for bronchial asthma evaluation by calling the computer program stored in the memory.

[0031] The electronic device may vary significantly due to different configurations or performance, and may include one or more processors (Central Processing Units, CPUs) and one or more memories, wherein the memories store at least one computer program, which is loaded and executed by the processor to implement the infectious disease risk prediction method based on the detection sharing network provided in the above method embodiment. The electronic device may also include other components for implementing the device functions. For example, the electronic device may also have components such as wired or wireless network interfaces and input and output interfaces for data input and output. This embodiment will not be described in detail here.

[0032] Example 4

[0033] In this embodiment, a computer-readable storage medium is provided, which stores a rewritable computer program. When the computer program is executed on a computer device, the computer device executes the above-mentioned nitric oxide data analysis method for bronchial asthma assessment.

[0034] The various embodiments of the present invention are described in a progressive manner, and reference can be made to the same or similar parts between the various embodiments. Each embodiment focuses on the differences from other embodiments.

[0035] The systems and media provided in the embodiments of the present invention correspond one-to-one with the methods. Therefore, the systems and media also have similar beneficial technical effects as their corresponding methods. Since the beneficial technical effects of the methods have been described in detail above, the beneficial technical effects of the systems and media will not be repeated here.

[0036] Those skilled in the art will appreciate that embodiments of the present invention may be implemented as methods, systems, or computer program products. Therefore, the present invention may be implemented entirely in hardware, entirely in software, or in a combination of software and hardware. Furthermore, the present invention may be implemented as a computer program product on one or more computer-usable storage media (e.g., disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0037] The present invention is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present invention. It should be understood that each process and / or block in the flowcharts and / or block diagrams, as well as combinations of processes and / or blocks in the flowcharts and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine. Execution of these instructions by the processor of the computer or other programmable data processing device can generate a device for implementing the functions specified in one or more processes in the flowcharts and / or one or more blocks in the block diagrams.

[0038] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to operate in a specific manner. In this way, the instructions stored in the computer-readable memory will produce an article of manufacture containing instruction means that can implement the functions specified by one or more flow charts and / or one or more blocks in the block diagrams.

[0039] In a typical configuration, a computing device includes one or more processors (CPUs), input / output interfaces, network interfaces, and memory. Memory may include non-permanent storage in computer-readable media, such as random access memory (RAM), and non-volatile memory, such as read-only memory (ROM) or flash memory (flash RAM). Memory is an example of a computer-readable medium.

[0040] Computer-readable media includes both permanent and non-permanent, removable and non-removable media that can be used to store information using any method or technology. Information can be computer-readable instructions, data structures, program modules, or other data. Examples of computer storage media include phase-change RAM (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), and other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, compact disc read-only memory (CD-ROM), digital versatile disc (DVD) or other optical storage, magnetic cassettes, magnetic tape, magnetic disk storage, or other magnetic storage devices, and any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer-readable media does not include transitory computer-readable media, such as modulated data signals and carrier waves.

Claims

1. A nitric oxide data analysis method for bronchial asthma assessment, characterized in that: The following steps are involved: S1. Collect patients' exhaled nitric oxide, alveolar nitric oxide, omics data and clinical information; S2. Clean, normalize and process the data for missing values ​​to obtain fused exhaled nitric oxide and alveolar nitric oxide, omics data and clinical information; S3. Construct the following features: inflammation level index, inflammation area entropy, inflammation synergy index, glucocorticoid response factor, infection-smoking synergy factor, and genetic-symptom distribution index; S4, divide the training set and test set; S5, using the interpretable boosting machine as the classifier and the decision tree as the base learner to iteratively train the model; S6. Using the trained model to output an initial assessment value of bronchial asthma; S7. Construct a final bronchial asthma assessment formula, substitute the initial bronchial asthma assessment value and various characteristics into the final bronchial asthma assessment formula, and obtain the bronchial asthma assessment grade.

2. The nitric oxide data analysis method for bronchial asthma assessment according to claim 1, characterized in that: The inflammation level index construction includes the following steps: using the nitric oxide data in the data set, exhaled nitric oxide and alveolar nitric oxide, to construct the following hierarchical features: inflammation level index, inflammation level index The derivation steps are as follows: standardize the exhaled nitric oxide by the reference exhaled nitric oxide concentration of healthy people, and perform power law transformation on the standardized exhaled nitric oxide to obtain the exhaled nitric oxide weight adjustment term Then, the alveolar nitric oxide was corrected by the eosinophil count, and the corrected alveolar nitric oxide was standardized to calculate the alveolar nitric oxide-eosinophil logarithmic enhancement factor. The inflammation level index evaluation formula is obtained by multiplying the exhaled nitric oxide weight adjustment term and the alveolar nitric oxide-eosinophil logarithmic enhancement term. The specific expression is: ,in, is the exhaled nitric oxide concentration, is the reference value of exhaled nitric oxide for healthy people, is alveolar nitric oxide, is the baseline value of alveolar nitric oxide, is the exhaled nitric oxide weight adjustment factor, Eosinophil count, is the correction factor for alveolar nitric oxide by eosinophils.

3. The nitric oxide data analysis method for bronchial asthma assessment according to claim 2, characterized in that: The entropy construction of the inflammation region includes the following steps: performing probability modeling on exhaled nitric oxide and alveolar nitric oxide to obtain , calculate the single point entropy of exhaled nitric oxide and alveolar nitric oxide respectively , combining the two single point entropies into the regional entropy , where k is any one of the exhaled nitric oxide concentration and the alveolar nitric oxide concentration.

4. The nitric oxide data analysis method for bronchial asthma assessment according to claim 3, characterized in that: The construction of the inflammation synergy index includes the following steps: multiplying the age attenuation term after sigmoid transformation with the Th2 basic synergy term and adding the allergy- Saturated interaction terms , ,in is the age-dependent attenuation factor, is the weight coefficient of allergy history, is a binary variable, is the total serum immune protein concentration.

5. The nitric oxide data analysis method for bronchial asthma assessment according to claim 4, characterized in that: The specific derivation process of the glucocorticoid response factor is as follows: Calculate the exhaled nitric oxide response attenuation term , multiplied by the drug dose and the inflammatory change rate to obtain the glucocorticoid response factor, which is expressed as ,in, is the daily dose of inhaled corticosteroids, for The rate of change of is the inflammation level index, calculated by sliding window: .

6. The nitric oxide data analysis method for bronchial asthma assessment according to claim 5, characterized in that: The specific derivation process of the infection-smoking synergistic influencing factor is as follows: smoke damages pulmonary vascular homeostasis through three mechanisms, and the risk assessment formula for smoking-induced pulmonary vascular damage is: ,in, The cumulative exposure to tobacco leads to the pathological transformation of pulmonary hemorrhage into asthma exacerbation. is the carboxyhemoglobin saturation, is the mean pulmonary artery pressure, It is a marker of pulmonary hypertension. The risk assessment formula for post-hemorrhagic inflammatory outbreak is: ,in The amount of bleeding, is the serum iron concentration.

7. The nitric oxide data analysis method for bronchial asthma assessment according to claim 6, characterized in that: The specific derivation process of the genetic-symptom distribution index is as follows: Multiplying symptoms The genetic risk switch is used to determine whether to include it, and finally a logarithmic transformation is performed. ,in, The core symptoms are scored from 0 to 3 according to severity. is the weight coefficient, is the predicted value based on age and height.

8. The nitric oxide data analysis method for bronchial asthma assessment according to claim 7, characterized in that: The final bronchial asthma assessment formula is ,in represents the sigmoid function, For the initial assessment of bronchial asthma, is the weight coefficient, This is the risk assessment result of post-bleeding inflammatory outbreak.

9. A nitric oxide data analysis system for bronchial asthma assessment, which is implemented based on the nitric oxide data analysis method for bronchial asthma assessment according to any one of claims 1 to 8, characterized in that: It includes the following modules: a data acquisition module, which is configured to obtain the patient's exhaled nitric oxide concentration, alveolar nitric oxide concentration, blood routine test data and proteomics data through standardized testing equipment, and simultaneously collect clinical information data sets, including age, smoking history, allergy history, lung function indicators and glucocorticoid dosage; The preprocessing module is configured to filter noise and remove outliers from the collected exhaled nitric oxide and alveolar nitric oxide and omics data, fill missing values ​​using multiple interpolation, and unify the dimensions of multi-source data through Z-score standardization; the fusion dimensionality reduction module is configured to perform feature-level splicing on the preprocessed exhaled nitric oxide and alveolar nitric oxide, omics data, and clinical information; A feature building module configured to perform the following calculations: generating an inflammation hierarchy index based on reference values ​​of exhaled nitric oxide and eosinophil counts in healthy individuals, calculating the entropy of inflammatory regions to quantify the spatial concentration of inflammation by modeling the probability distribution of exhaled nitric oxide concentrations; and generating a drug response factor by combining the daily glucocorticoid dose and the rate of change of exhaled nitric oxide. The infection-smoking synergistic influencing factor was calculated by integrating the number of years of smoking, respiratory infection status, and alveolar nitric oxide constraints. A genetic-symptom distribution index was constructed by weighting the lung function deficit term with the symptoms. The model training module was configured as follows: the data was divided into training and test sets in a 7:3 ratio; the model was iteratively trained using the interpretable boosting machine (EBM) as the classifier and the decision tree as the base learner via the gradient boosting algorithm; grid search was used to optimize hyperparameters, and cross-validation was performed using the Matthews correlation coefficient (MCC) as the evaluation metric. The evaluation output module is configured to input the fusion features of the patient to be evaluated into the trained EBM model and output a bronchial asthma risk score.

Images