Asthma risk alerting system and method based on respiratory monitoring

Abstract

This application relates to the field of medical data processing technology, specifically to an asthma risk early warning system and method based on respiratory monitoring. The method includes: collecting physiological and environmental data from asthma patients; obtaining the asthma response sensitivity of asthma patients to various types of environmental data based on the physiological and environmental data; constructing a long-term linked environmental data type set based on the environmental data, and obtaining the influence degree of multiple linked environmental data patterns of asthma patients based on the long-term linked environmental data type set; obtaining the multi-dimensional environmental risk influence coefficient of asthma patients based on the asthma response sensitivity and the influence degree of multiple linked environmental data patterns; inputting the multi-dimensional environmental risk influence coefficient and the physiological data of asthma patients into a pre-trained neural network model, and outputting the asthma risk early warning result for the asthma patients. This can achieve more accurate and personalized asthma risk early warning, and improve the reliability of the system's asthma risk early warning under complex environmental changes.

Description

Technical Field

[0001] This application belongs to the field of medical data processing technology, specifically relating to an asthma risk early warning system and method based on respiratory monitoring. Background Technology

[0002] Asthma is a chronic respiratory disease whose sudden acute attacks seriously threaten patients' lives. Traditional early warning methods rely on threshold judgments or time-series analysis of physiological parameters (such as airway resistance and respiratory rate), but neglect the dynamic influence of environmental factors (such as PM2.5, temperature, and humidity). Existing technologies have failed to deeply explore the coupling mechanism between the environment and physiology, resulting in insufficient timeliness and specificity of early warnings and a high false alarm rate. Therefore, there is an urgent need for a solution that can integrate multi-dimensional data to achieve personalized and accurate early warnings. Summary of the Invention

[0003] To address the aforementioned issues, this application provides an asthma risk early warning system and method based on respiratory monitoring.

[0004] According to a first aspect of the embodiments of this application, an asthma risk early warning method based on respiratory monitoring is provided, the method comprising: Collect physiological and environmental data from asthma patients; Based on the physiological data and the environmental data, the asthma response sensitivity of the asthma patient to various types of environmental data is obtained; Based on the environmental data, a long-term linked environmental data type set is constructed, and based on the long-term linked environmental data type set, the influence degree of the multi-linked environmental data pattern of the asthma patient is obtained. Based on the asthma response sensitivity and the influence of the multi-linked environmental data pattern, the multi-dimensional environmental risk impact coefficient of the asthma patient is obtained. The multidimensional environmental risk impact coefficient and the physiological data of the asthma patient are input into a pre-trained neural network model, which outputs the asthma risk warning result of the asthma patient.

[0005] In one embodiment, the physiological data includes at least one of peak expiratory flow rate, heart rate, blood oxygen saturation, or respiratory rate; The environmental data includes at least one of PM2.5 levels, nitrogen oxides, temperature, humidity, air pressure, wind speed, or precipitation.

[0006] In one implementation, acquiring the asthma response sensitivity of the asthma patient to multiple types of environmental data includes: The environmental data is clustered into multiple clusters based on clustering methods, and each cluster contains multi-dimensional data analysis intervals under similar environmental data. For each cluster, the relative variability of environmental data and the variability of peak expiratory flow rate for the asthma patients were obtained. Based on the relative variability fluctuation and peak expiratory flow fluctuation, the asthma response sensitivity of the asthma patients to each type of environmental data is obtained.

[0007] In one implementation, the relative variability fluctuation is obtained by calculating and normalizing the standard deviation of the same type of environmental data of the asthmatic patient; the peak expiratory flow rate fluctuation is obtained by calculating and normalizing the standard deviation of the peak expiratory flow rate data of the asthmatic patient.

[0008] In one implementation, obtaining the asthma response sensitivity of the asthma patient to each type of environmental data based on the relative variability fluctuation and peak expiratory flow variability includes: Based on the environmental data of each type of asthma patient, the peak expiratory flow fluctuation performance under each cluster is multiplied by the relative variability fluctuation to obtain the first data; The first data of all clusters corresponding to each type of environmental data of the asthma patient are summed to obtain the second data; The second data is normalized to obtain the asthma response sensitivity of the asthma patients to each type of environmental data.

[0009] In one implementation, the step of constructing a long-term linked environmental data type set based on the environmental data, and obtaining the influence degree of the multi-linked environmental data patterns of the asthma patient based on the long-term linked environmental data type set, includes: Based on the daily environmental data of the asthma patients, the degree of continuous change of each type of environmental data for each day is obtained; Based on the continuous transformation degree, a long-term continuous transformation curve for each type of environmental data is constructed. Obtain the correlation between the long-term continuous transformation curves of any two environmental data, filter out environmental data pairs with a correlation greater than a preset threshold, and construct the long-term linkage environmental data type set based on the data type of the environmental data pairs; Based on the correlation between the environmental data corresponding to the long-term linked environmental data type set, the influence of the multi-linked environmental data pattern of the asthma patient is obtained.

[0010] In one implementation, obtaining the multi-dimensional environmental risk impact coefficient of the asthma patient based on the asthma response sensitivity and the influence of the multi-linked environmental data pattern includes: Based on the physiological data, the relative abnormality of the peak expiratory flow rate of the asthma patients was obtained; Based on the influence of the multi-linkage environment data pattern, the anomalous influence of the multi-linkage environment data pattern of the asthma patient is obtained; Based on the asthma response sensitivity, the degree of abnormality in single-type environmental data of the asthma patient is obtained; Based on the relative abnormality of the peak expiratory flow rate, the abnormal impact of the multi-linked environmental data pattern, and the abnormal performance of the single-type environmental data, the multi-dimensional environmental risk impact coefficient of the asthma patient is obtained.

[0011] In one embodiment, the method further includes: Based on the output of the neural network model, an asthma risk warning signal is generated and sent to the asthma patient or the asthma patient's attending physician via a smart device.

[0012] In one implementation, the neural network model is trained using a training set, which includes multiple training samples from the asthma patient. The training samples include the asthma patient's physiological data, multi-dimensional environmental risk impact coefficients, and corresponding asthma risk warning levels.

[0013] According to a second aspect of the embodiments of this application, an asthma risk early warning system based on respiratory monitoring is provided, the system including a server, the server comprising: A memory on which computer programs are stored; A processor for executing the computer program in the memory to implement the steps of the method of any one of the first aspects.

[0014] The embodiments of this application have the following beneficial effects: This application provides an asthma risk early warning method based on respiratory monitoring. The method includes: collecting physiological and environmental data of asthma patients; obtaining the asthma response sensitivity of asthma patients to various types of environmental data based on the physiological and environmental data; constructing a long-term linked environmental data type set based on the environmental data, and obtaining the influence degree of multiple linked environmental data patterns of asthma patients based on the long-term linked environmental data type set; obtaining the multi-dimensional environmental risk influence coefficient of asthma patients based on the asthma response sensitivity and the influence degree of multiple linked environmental data patterns; inputting the multi-dimensional environmental risk influence coefficient and the physiological data of asthma patients into a pre-trained neural network model, and outputting the asthma risk early warning result of asthma patients. This application embodiment constructs a long-term linked environmental dataset and influence index; finally, by combining the patient's real-time peak expiratory flow abnormality and the abnormal performance of environmental factors, it dynamically fuses and generates a multi-dimensional environmental risk influence coefficient, and inputs it along with traditional respiratory data into a neural network model for training and prediction, thereby achieving more accurate and personalized asthma risk early warning and improving the reliability of the system's asthma risk early warning under complex environmental changes.

[0015] Other features and advantages of this application will be described in detail in the following detailed description section. Attached Figure Description

[0016] To more clearly illustrate the implementation schemes of this application, the accompanying drawings used in the implementation schemes will be briefly introduced below. It should be understood that the accompanying drawings only show some implementation schemes of this application and should not be regarded as a limitation of the scope. For those skilled in the art, other related drawings can be obtained from the accompanying drawings without creative effort.

[0017] Figure 1 This is a flowchart illustrating an asthma risk warning method based on respiratory monitoring, according to an exemplary embodiment.

[0018] Figure 2 This is a flowchart illustrating a method for obtaining the asthma response sensitivity of asthma patients to various types of environmental data, according to an exemplary embodiment.

[0019] Figure 3 This is a flowchart illustrating a method for obtaining relative variability fluctuations and peak expiratory flow variability in various types of environmental data of asthma patients, according to an exemplary embodiment.

[0020] Figure 4 This is a flowchart illustrating a method for obtaining the asthma response sensitivity of asthma patients to each type of environmental data based on relative variability fluctuation and peak expiratory flow variability, according to an exemplary embodiment.

[0021] Figure 5 This is a flowchart illustrating, according to an exemplary embodiment, a method for constructing a long-term linked environmental data type set based on environmental data, and obtaining the influence of multiple linked environmental data patterns of asthma patients based on the long-term linked environmental data type set.

[0022] Figure 6 This is a flowchart illustrating a method for obtaining the multi-dimensional environmental risk impact coefficient of asthma patients based on asthma response sensitivity and the impact of multi-linked environmental data patterns, according to an exemplary embodiment.

[0023] Figure 7 This is a flowchart illustrating yet another asthma risk warning method based on respiratory monitoring, according to an exemplary embodiment.

[0024] Figure 8 This is a block diagram illustrating an asthma risk warning system based on respiratory monitoring, according to an exemplary embodiment.

[0025] Figure 9This is a block diagram illustrating a server according to an exemplary embodiment. Detailed Implementation

[0026] To clearly illustrate the technical features of this solution, the following detailed description, in conjunction with specific implementation methods and accompanying drawings, will provide a comprehensive explanation of this application.

[0027] Embodiments of this application will now be described in more detail with reference to the accompanying drawings. While some embodiments of this application are shown in the drawings, it should be understood that this application can be implemented in various forms and should not be construed as limited to the embodiments set forth herein. Rather, these embodiments are provided to provide a more thorough and complete understanding of this application. It should be understood that the drawings and embodiments of this application are for illustrative purposes only and are not intended to limit the scope of protection of this application.

[0028] It should be understood that the steps described in the method embodiments of this application may be performed in different orders and / or in parallel. Furthermore, the method embodiments may include additional steps and / or omit the steps shown. The scope of this application is not limited in this respect.

[0029] The term "comprising" and its variations as used herein are open-ended inclusions, meaning "including but not limited to". The term "based on" means "at least partially based on". The term "one embodiment" means "at least one embodiment"; the term "another embodiment" means "at least one additional embodiment"; the term "some embodiments" means "at least some embodiments". Definitions of other terms will be given in the description below.

[0030] It should be noted that the concepts of "first" and "second" mentioned in this application are only used to distinguish different devices, modules or units, and are not used to limit the order of functions performed by these devices, modules or units or their interdependencies.

[0031] It should be noted that the terms "one" and "multiple" used in this application are illustrative rather than restrictive. Those skilled in the art should understand that, unless explicitly stated in the context, they should be interpreted as "one or more". In the description of this application, unless otherwise stated, "multiple" refers to two or more than two, and other quantifiers are similar; "at least one item", "one item or multiple items", or similar expressions refer to any combination of these items, including any combination of single or plural items. For example, at least one item 'a' can represent any number of 'a's; as another example, one or more of a, b, and c can represent: a, b, c, ab, ac, bc, or abc, where a, b, and c can be single or multiple; "and / or" describes the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A alone, A and B simultaneously, and B alone, where A and B can be singular or plural.

[0032] Although operations or steps are described in a specific order in the accompanying drawings in the embodiments of this application, this should not be construed as requiring these operations or steps to be performed in the specific order or serial order shown, or requiring all of the shown operations or steps to be performed to obtain the desired result. In the embodiments of this application, these operations or steps may be performed serially; they may be performed in parallel; or a portion of these operations or steps may be performed.

[0033] Meanwhile, it is understood that the data involved in this technical solution (including but not limited to the data itself, the acquisition or use of the data) shall comply with the requirements of relevant laws, regulations and related provisions.

[0034] First, the application scenario of this application will be explained. Traditional asthma risk warning methods mainly rely on the monitoring and analysis of multiple respiratory physiological parameters, such as airway resistance, respiratory rate, and blood oxygen saturation, to assess and predict the risk of acute asthma attacks by identifying abnormal changes. However, the occurrence and development of asthma is a complex process involving multiple interacting factors. It is not only closely related to individual physiological state, but environmental factors also play a crucial role. For example, increased concentrations of pollutants such as PM2.5 and pollen in the air, as well as sudden changes in temperature and humidity, often become key factors inducing the aggravation of asthma symptoms or acute attacks. Most existing warning models focus on threshold judgments or time-series change analysis of physiological parameters, failing to effectively incorporate environmental variables and lacking in-depth exploration of the dynamic coupling mechanism between environment and physiology. Therefore, warning systems built solely based on physiological parameters have significant limitations in terms of specificity and timeliness. Especially when the environment changes drastically, the model may not be able to provide sufficient advance warning or may produce a high false alarm rate, thereby reducing clinical applicability and user compliance.

[0035] In view of this, this application provides an asthma risk early warning system and method based on respiratory monitoring, aiming to solve the above problems. The application will be described below with reference to specific embodiments.

[0036] Figure 1 This is a flowchart illustrating an asthma risk early warning method based on respiratory monitoring, according to an exemplary embodiment. Figure 1 As shown in the figure, this application provides an asthma risk early warning method based on respiratory monitoring, which may include the following steps: In step S10, physiological and environmental data of asthma patients are collected.

[0037] In this step, physiological and environmental data of asthma patients are collected. For example, fluctuations in air pollutants caused by environmental changes are a significant factor affecting the health of asthma patients. Major air pollutants, such as PM2.5 and nitrogen oxides, are the main external triggers for asthma attacks. Especially during peak pollution periods, respiratory symptoms in asthma patients may worsen, and the incidence and frequency of acute exacerbations of asthma increase significantly with long-term exposure to polluted environments. Furthermore, the impact of the environment on asthma is not limited to air pollutants; other meteorological conditions such as temperature, humidity, air pressure, wind speed, and precipitation also significantly affect the patient's condition. Therefore, to optimize asthma risk warning methods based on respiratory monitoring, it is necessary to combine the impact of environmental factors on asthma and provide a more comprehensive health management solution: using portable peak flow meters to measure peak expiratory flow data at regular intervals, and using devices such as smart bracelets to collect respiratory data such as heart rate, blood oxygen saturation, and respiratory rate; simultaneously, using portable personal environmental sensors to acquire and detect air pollutants such as PM2.5 and nitrogen oxides, as well as environmental temperature and humidity data in real time.

[0038] In step S20, based on the physiological data and the environmental data, the asthma response sensitivity of the asthma patient to multiple types of environmental data is obtained.

[0039] In this step, based on physiological and environmental data, the asthma response sensitivity of asthma patients to various types of environmental data is obtained. For example, environmental data can first be clustered into multiple clusters using clustering methods. Each cluster contains multi-dimensional data analysis intervals for similar environmental data. Then, for each cluster, the relative variability fluctuation and peak expiratory flow fluctuation of asthma patients for various types of environmental data are obtained. Finally, based on the relative variability fluctuation and peak expiratory flow fluctuation, the asthma response sensitivity of asthma patients to each type of environmental data is obtained.

[0040] In step S30, based on the environmental data, a long-term linked environmental data type set is constructed, and based on the long-term linked environmental data type set, the influence degree of the multi-linked environmental data pattern of the asthma patient is obtained.

[0041] In this step, a long-term linked environmental data type set is constructed based on environmental data, and the influence of multiple linked environmental data patterns in asthma patients is obtained based on this set. For example, based on the historical daily environmental data of asthma patients, the continuous variability of each type of environmental data for each day can be obtained. Then, based on the continuous variability, a long-term continuous variability curve for each type of environmental data can be constructed. Next, the correlation between the long-term continuous variability curves of any two types of environmental data can be obtained, and environmental data pairs with a correlation greater than a preset threshold are selected. A long-term linked environmental data type set is constructed based on the data type of these environmental data pairs. Finally, based on the correlation of the environmental data corresponding to the long-term linked environmental data type set, the influence of multiple linked environmental data patterns in asthma patients is obtained.

[0042] In step S40, based on the asthma response sensitivity and the influence of the multi-linked environmental data pattern, the multi-dimensional environmental risk influence coefficient of the asthma patient is obtained.

[0043] In this step, a multi-dimensional environmental risk impact coefficient for asthma patients is obtained based on asthma response sensitivity and the influence of multi-linked environmental data patterns. For example, firstly, based on physiological data, the relative abnormality of peak expiratory flow rate in asthma patients can be obtained; then, based on the influence of multi-linked environmental data patterns, the anomalous influence of these patterns can be obtained; next, based on asthma response sensitivity, the anomalous performance of a single type of environmental data can be obtained; finally, based on the relative abnormality of peak expiratory flow rate, the anomalous influence of multi-linked environmental data patterns, and the anomalous performance of a single type of environmental data, the multi-dimensional environmental risk impact coefficient for asthma patients is obtained.

[0044] In step S50, the multi-dimensional environmental risk impact coefficient and the physiological data of the asthma patient are input into a pre-trained neural network model, and the asthma risk warning result of the asthma patient is output.

[0045] In this step, multi-dimensional environmental risk impact coefficients and physiological data of asthma patients are input into a pre-trained neural network model, which outputs asthma risk warning results for asthma patients.

[0046] In one possible implementation, the physiological data includes at least one of peak expiratory flow, heart rate, blood oxygen saturation, or respiratory rate.

[0047] Environmental data include at least one of PM2.5 levels, nitrogen oxides, temperature, humidity, air pressure, wind speed, or precipitation.

[0048] In one possible implementation, the neural network model is trained using a training set that includes multiple training samples from asthma patients. These training samples include the patients' physiological data, multi-dimensional environmental risk impact coefficients, and corresponding asthma risk warning levels. For example, a large dataset (over 30,000) of asthma patients from different periods can be collected. From these datasets, the patients' respiratory data and corresponding multi-dimensional environmental risk impact coefficients are extracted sequentially, and manual evaluation is performed to determine the corresponding asthma risk warning level. After processing all datasets, the training and validation sets are divided in a 7:3 ratio. The neural network is trained using the training set samples, with the cross-entropy function as the loss function. Gradient descent is used to train until the loss function converges. The robustness of the training results is then verified using the validation set, resulting in the trained neural network.

[0049] Figure 2 This is a flowchart illustrating a method for obtaining the asthma response sensitivity of asthma patients to various types of environmental data, according to an exemplary embodiment. Figure 2 As shown, obtaining the asthma response sensitivity of the asthma patient to various types of environmental data may include the following steps: In step S201, the environmental data is clustered into multiple clusters based on a clustering method, and each cluster contains multi-dimensional data analysis intervals under similar environmental data.

[0050] In this step, environmental data is clustered into multiple clusters based on clustering methods. Each cluster contains multi-dimensional data analysis intervals for similar environmental data. For example, due to individual differences among asthma patients, even under the influence of the same or similar environmental factors, different patients may have different sensitivities to these factors. This difference may stem from various factors such as genetic factors, differences in the immune system, underlying health conditions, and lifestyle habits. Specifically, some patients may exhibit stronger symptom responses and be at higher asthma risk when faced with high concentrations of PM2.5, nitrogen oxides, or extreme temperature and humidity changes, while other patients may be more tolerant of these environmental changes and be at lower asthma risk. Therefore, individual differences in environmental adaptability and physiological characteristics are key factors determining the level of asthma attack risk when patients face different environmental factors. Typically, respiratory and environmental data collected from asthma patients are uniformly aggregated in the hospital's electronic medical record system, where relevant data for several asthma patients are obtained. The collected patient environmental data can be divided into samples (each sample corresponds to environmental data collected half an hour before and after a peak expiratory flow). By calculating the Euclidean distance between any two samples, the DBSCAN clustering method is used to group samples with similar environmental characteristics into the same cluster, forming multiple multi-dimensional data analysis interval clusters containing similar environmental data.

[0051] In step S202, for each cluster, the relative variability fluctuation and peak expiratory flow fluctuation of the various types of environmental data of the asthma patients are obtained.

[0052] In this step, for each cluster, the relative variability and peak expiratory flow variability of various types of environmental data for asthma patients are obtained. For example, the relative variability is obtained by calculating and normalizing (e.g., norm normalization) the standard deviation of the same type of environmental data for asthma patients; the peak expiratory flow variability is obtained by calculating and normalizing (e.g., norm normalization) the standard deviation of the peak expiratory flow data for asthma patients.

[0053] In step S203, based on the relative variability fluctuation degree and peak expiratory flow fluctuation performance degree, the asthma response sensitivity of the asthma patient to each type of environmental data is obtained.

[0054] In this step, the asthma response sensitivity of asthma patients to each type of environmental data is obtained based on the relative variability fluctuation and the peak expiratory flow rate fluctuation performance. For example, firstly, based on each type of environmental data for asthma patients, the peak expiratory flow rate fluctuation performance under each cluster is multiplied by the relative variability fluctuation to obtain the first data. Then, the first data for all clusters corresponding to each type of environmental data for asthma patients are summed to obtain the second data. Finally, the second data is normalized (e.g., Softmax normalization) to obtain the asthma response sensitivity of asthma patients to each type of environmental data.

[0055] Figure 3 This is a flowchart illustrating a method for acquiring relative variability in the fluctuation of various types of environmental data and peak expiratory flow variability in asthma patients, according to an exemplary embodiment. Figure 3 As shown, the acquisition of relative variability fluctuations and peak expiratory flow rate fluctuations in various types of environmental data for the asthma patient may include the following steps: In step S2021, the relative variability fluctuation is obtained by calculating and normalizing the standard deviation of the same type of environmental data for the asthma patient. The peak expiratory flow rate fluctuation is obtained by calculating and normalizing the standard deviation of the peak expiratory flow rate data for the asthma patient.

[0056] In this step, the relative variability fluctuation is obtained by calculating and normalizing the standard deviation of environmental data of the same type for asthma patients. For example, when calculating the relative variability fluctuation, all sample values ​​of environmental data of the same type for the same patient in the same cluster can be extracted first (e.g., PM2.5 data of all samples of a patient in cluster 1), and the standard deviation of these sample values ​​can be calculated to reflect the fluctuation range of the environmental data under similar environments; then, Softmax normalization is used to process the standard deviation to eliminate the difference in the units of different environmental data (e.g., the unit difference between PM2.5 and temperature), so as to obtain a relative variability fluctuation that can be compared horizontally.

[0057] Peak expiratory flow rate (PEFR) variability is determined by calculating and normalizing the standard deviation of PEFR data from asthma patients. For example, to calculate PEFR variability, PEFR data from all samples within the same cluster for the same patient can be extracted, their standard deviations calculated to reflect physiological fluctuations, and then normalized using norm to eliminate individual differences in baseline PEFR among patients (such as differences in baseline values ​​due to height and age), resulting in a standardized PEFR variability.

[0058] A unified normalized calculation method makes the fluctuation indicators of different types of environmental data and the physiological fluctuation indicators of different patients comparable, avoiding analytical bias caused by differences in the attributes of the data itself, and laying the foundation for the accurate calculation of asthma response sensitivity in the future.

[0059] Figure 4 This is a flowchart illustrating a method for obtaining the asthma response sensitivity of asthma patients to each type of environmental data based on relative variability fluctuation and peak expiratory flow variability, according to an exemplary embodiment. Figure 4 As shown, obtaining the asthma response sensitivity of the asthma patient to each type of environmental data based on the relative variability fluctuation and peak expiratory flow fluctuation performance may include the following steps: In step S2031, based on each type of environmental data of the asthma patient, the peak expiratory flow rate fluctuation performance under each cluster is multiplied by the relative variability fluctuation to obtain the first data.

[0060] In this step, based on each type of environmental data for asthma patients, the peak expiratory flow rate fluctuation performance under each cluster is multiplied by the relative variability fluctuation to obtain the first data. For example, for each type of environmental data for the patient (e.g., temperature, PM2.5), all clusters corresponding to that environmental data can be processed one by one: within each cluster, the peak expiratory flow rate fluctuation performance of the patient under that cluster is multiplied by the relative variability fluctuation of that type of environmental data to obtain the impact value of that environmental data on the patient's asthma response within that cluster (the first data).

[0061] In step S2032, the first data of all clusters corresponding to each type of environmental data of the asthma patient are summed to obtain the second data.

[0062] In this step, the first data for all clusters corresponding to each type of environmental data for asthma patients are summed to obtain the second data. For example, after all clusters have been processed, the second data can be obtained by summing all the first data for that type of environmental data, thus integrating environmental impact information across multiple clusters.

[0063] In step S2033, the second data is normalized to obtain the asthma response sensitivity of the asthma patient to each type of environmental data.

[0064] In this step, the second data is normalized to obtain the asthma response sensitivity of asthma patients to each type of environmental data. For example, the second data can be subjected to Softmax normalization to eliminate the range differences in the accumulated results between different environmental data, thus obtaining the patient's asthma response sensitivity to that type of environmental data. For instance, for the PM2.5 environmental data type, its first data in cluster 1, cluster 2, and cluster 3 are calculated separately, summed, and then normalized (e.g., norm normalization) to finally obtain the patient's asthma response sensitivity to PM2.5.

[0065] By accumulating and normalizing multi-cluster data, the analysis results from different similar environmental scenarios are integrated, avoiding the limitations of single-cluster data. This makes the sensitivity assessment more comprehensive and objective, and can accurately reflect the patient's true sensitivity to each type of environmental data.

[0066] Figure 5 This is a flowchart illustrating, according to an exemplary embodiment, a method for constructing a long-term linked environmental data type set based on environmental data, and obtaining the influence of multiple linked environmental data patterns in asthma patients based on this long-term linked environmental data type set. Figure 5As shown, the step of constructing a long-term linked environmental data type set based on the environmental data, and obtaining the influence degree of the multi-linked environmental data patterns of the asthma patient based on the long-term linked environmental data type set, may include the following steps: In step S301, based on the historical daily environmental data of the asthma patient, the degree of continuous change of each type of environmental data for each day is obtained.

[0067] In this step, based on the historical daily environmental data of asthma patients, the degree of continuous variability for each type of environmental data is obtained for each day. For example, the historical mean of each type of environmental data can be used as a reference assessment value. The difference between two adjacent environmental data collections within a day is taken and the absolute value is compared with the reference assessment value (prior value) to obtain the degree of variability between the two adjacent data collections. The sum of all adjacent variability values ​​for the day is then calculated and the mean is obtained, which is the degree of continuous variability for that environmental data for that day.

[0068] In step S302, based on the continuous transformation degree, a long-term continuous transformation curve for each type of environmental data is constructed.

[0069] In this step, a long-term continuous transformation curve is constructed for each type of environmental data based on the degree of continuous transformation. For example, the long-term continuous transformation curve for each type of environmental data can be constructed using the least squares method based on the degree of continuous transformation each day, intuitively presenting the long-term trend of environmental data changes.

[0070] In step S303, the correlation between the long-term continuous transformation curves of any two environmental data is obtained, environmental data pairs with a correlation greater than a preset threshold are selected, and the long-term linkage environmental data type set is constructed based on the data type of the environmental data pairs.

[0071] In this step, the correlation between the long-term continuous transformation curves of any two environmental data sets is obtained. Environmental data pairs with a correlation greater than a preset threshold are selected, and a long-term linked environmental data type set is constructed based on the data type of the environmental data pairs. For example, the correlation between the long-term continuous transformation curves of any two environmental data sets can be calculated using the Pearson correlation coefficient. Environmental data pairs with a correlation greater than a preset threshold (such as 0.65) are selected, and a long-term linked environmental data type set is constructed based on the type of these data pairs.

[0072] In step S304, the influence of the multi-linkage environmental data pattern of the asthma patient is obtained based on the correlation of the environmental data corresponding to the long-term linkage environmental data type set.

[0073] In this step, the influence of multiple linked environmental data patterns on asthma patients is obtained based on the correlation between environmental data corresponding to the long-term linked environmental data set. For example, the correlation of all environmental data pairs within the set can be summed and averaged. This average value is the influence of the patient's multiple linked environmental data patterns. The higher the average value, the stronger the influence of the linked environment on the patient.

[0074] This process effectively captures the inherent correlations between environmental data, identifies environmental combinations that may have a combined impact on patients' asthma, and is consistent with the fact that asthma attacks are often triggered by multiple environmental factors in reality, providing key linked environmental information for subsequent comprehensive risk assessment.

[0075] Figure 6 This is a flowchart illustrating a method for obtaining multi-dimensional environmental risk impact coefficients for asthma patients based on asthma response sensitivity and the impact of multi-linked environmental data patterns, according to an exemplary embodiment. Figure 6 As shown, obtaining the multi-dimensional environmental risk impact coefficient of the asthma patient based on the asthma response sensitivity and the influence of the multi-linked environmental data pattern may include the following steps: In step S401, based on the physiological data, the relative abnormality of the peak expiratory flow rate of the asthma patient is obtained.

[0076] In this step, based on physiological data, the relative abnormality W of peak expiratory flow rate for asthma patients is obtained. For example, the average of the patient's historical peak expiratory flow rates can be used as a reference. The current peak expiratory flow rate is compared to the reference data, and then normalized inversely. That is, the reciprocal of the current peak expiratory flow rate compared to the reference data is normalized to obtain the relative abnormality W. The greater the deviation of the current value from the reference value, the larger W is. W reflects the degree of abnormality in the patient's current respiratory function.

[0077] In step S402, based on the influence degree of the multi-linkage environmental data pattern, the anomalous influence degree of the multi-linkage environmental data pattern of the asthma patient is obtained.

[0078] In this step, based on the impact of the multi-linkage environment data pattern, the anomalous impact degree E of the multi-linkage environment data pattern for asthma patients is obtained. For example, based on a long-term multi-linkage environment data set, the impact degree of the daily multi-linkage environment data pattern can be calculated. This is then compared to the impact degree of the patient's long-term multi-linkage environment data pattern, and after inverse proportional normalization (i.e., normalizing the reciprocal of the ratio of the daily multi-linkage environment data pattern's impact degree to the patient's long-term multi-linkage environment data pattern's impact degree), the anomalous impact degree E of the multi-linkage environment data pattern is obtained. A larger E indicates a greater difference between the daily multi-linkage environment and the long-term norm.

[0079] In step S403, based on the asthma response sensitivity, the abnormality of the single-type environmental data of the asthma patient is obtained.

[0080] In this step, based on asthma response sensitivity, the anomalousness R of a single type of environmental data for asthma patients is obtained. For example, the historical mean of each type of environmental data can be used as a reference. The current environmental data is compared to the reference, and normalized using norm to obtain the environmental anomaly. The anomaly of each type of environmental data is multiplied by its corresponding asthma response sensitivity, summed, and then normalized again to obtain the anomalousness R of the single type of environmental data. R reflects the comprehensive impact of anomalous environmental data on the patient.

[0081] In step S404, the multi-dimensional environmental risk impact coefficient of the asthma patient is obtained based on the relative abnormality of the peak expiratory flow, the abnormal impact of the multi-linked environmental data pattern, and the abnormal performance of the single type of environmental data.

[0082] In this step, the multidimensional environmental risk impact coefficient Q for asthma patients is obtained based on the relative abnormality W of peak expiratory flow, the anomalous impact E of multi-linked environmental data patterns, and the anomalous performance R of single-type environmental data. For example, the multidimensional environmental risk impact coefficient Q for asthma patients can be obtained by the following formula: When W is high, the influence of E is emphasized; when W is low, the influence of R is emphasized, achieving a dynamic integration of physiological and environmental abnormalities. By comprehensively considering multi-dimensional indicators of physiological state and environmental factors, the risks faced by patients are fully assessed, taking into account both the impact of individual environmental factors and the interconnected effects of environmental data, ensuring that the risk coefficient accurately reflects the patient's real-time comprehensive risk level.

[0083] Figure 7 This is a flowchart illustrating yet another asthma risk warning method based on respiratory monitoring, according to an exemplary embodiment. Figure 7 As shown, the method may further include the following steps: In step S60, based on the output of the neural network model, an asthma risk warning signal is generated and sent to the asthma patient or the asthma patient's attending physician via a smart device.

[0084] In this step, based on the output of the neural network model, an asthma risk warning signal is generated and sent to the asthma patient or their attending physician via a smart device. For example, after the neural network model outputs the asthma risk warning result, the system generates a corresponding asthma risk warning signal based on the risk level of the warning result (e.g., high risk, medium risk, low risk): a high-risk signal corresponds to an emergency alert, a medium-risk signal corresponds to a warning, and a low-risk signal corresponds to a routine notification. Subsequently, the system pushes the warning information to the patient or their attending physician via a smart device (e.g., the patient's smartphone, smart bracelet, or the attending physician's office terminal). The information includes the current risk level, possible influencing factors (e.g., abnormal PM2.5 concentration, temperature and humidity changes), and recommended measures (e.g., reducing outdoor activities, using medication).

[0085] Timely early warning signals allow patients to take protective measures in advance to avoid exposure to high-risk environments. Attending physicians can also adjust treatment plans in a timely manner based on the warning information, effectively reducing the health risks caused by asthma attacks and improving the timeliness and proactivity of asthma management.

[0086] This application also provides a computer-readable storage medium having stored thereon computer program instructions that, when executed by a processor, implement the steps of the asthma risk warning method based on respiratory monitoring provided in this application.

[0087] Figure 8 This is a block diagram illustrating an asthma risk early warning system based on respiratory monitoring, according to an exemplary embodiment. Figure 8 As shown in the figure, this application provides an asthma risk early warning system 800 based on respiratory monitoring, including a server 900.

[0088] Figure 9 This is a block diagram illustrating a server according to an exemplary embodiment. (Refer to...) Figure 9 The server 900 includes a processor 922, which further includes one or more processors, and memory resources represented by memory 932 for storing instructions, such as applications, that can be executed by the processor 922. The applications stored in memory 932 may include one or more modules, each corresponding to a set of instructions. Furthermore, the processor 922 is configured to execute instructions to perform the aforementioned asthma risk warning method based on respiratory monitoring.

[0089] Server 900 may also include a power supply component 926 configured to perform power management of server 900, a communication component 950 configured to connect server 900 to a network, and an input / output interface 958. Server 900 can operate on an operating system stored in memory 932.

[0090] In another exemplary embodiment, a computer program product is also provided, comprising a computer program executable by a programmable electronic device, the computer program having a code portion for performing the above-described respiratory monitoring-based asthma risk warning method when executed by the programmable electronic device.

[0091] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of this application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these modifications and improvements all fall within the protection scope of this application.

Claims

1. A method for early warning of asthma risk based on respiratory monitoring, characterized in that, The method includes: Collect physiological and environmental data from asthma patients; Based on the physiological data and the environmental data, the asthma response sensitivity of the asthma patient to various types of environmental data is obtained; Based on the environmental data, a long-term linked environmental data type set is constructed, and based on the long-term linked environmental data type set, the influence degree of the multi-linked environmental data pattern of the asthma patient is obtained. Based on the asthma response sensitivity and the influence of the multi-linked environmental data pattern, the multi-dimensional environmental risk impact coefficient of the asthma patient is obtained. The multidimensional environmental risk impact coefficient and the physiological data of the asthma patient are input into a pre-trained neural network model, which outputs the asthma risk warning result of the asthma patient.

2. The asthma risk early warning method based on respiratory monitoring according to claim 1, characterized in that, The physiological data include at least one of peak expiratory flow, heart rate, blood oxygen saturation, or respiratory rate; The environmental data includes at least one of PM2.5 levels, nitrogen oxides, temperature, humidity, air pressure, wind speed, or precipitation.

3. The asthma risk early warning method based on respiratory monitoring according to claim 1, characterized in that, The acquisition of the asthma response sensitivity of the asthma patients to various types of environmental data includes: The environmental data is clustered into multiple clusters based on clustering methods, and each cluster contains multi-dimensional data analysis intervals under similar environmental data. For each cluster, the relative variability of environmental data and the variability of peak expiratory flow rate for the asthma patients were obtained. Based on the relative variability fluctuation and peak expiratory flow fluctuation, the asthma response sensitivity of the asthma patients to each type of environmental data is obtained.

4. The asthma risk early warning method based on respiratory monitoring according to claim 3, characterized in that, The acquisition of relative variability fluctuations and peak expiratory flow rate fluctuations in various types of environmental data for the asthma patients includes: The relative variability fluctuation is obtained by calculating and normalizing the standard deviation of the same type of environmental data of the asthma patients; the peak expiratory flow fluctuation is obtained by calculating and normalizing the standard deviation of the peak expiratory flow data of the asthma patients.

5. The asthma risk early warning method based on respiratory monitoring according to claim 3, characterized in that, The asthma response sensitivity of the asthma patient to each type of environmental data is obtained based on the relative variability fluctuation and peak expiratory flow variability, including: Based on the environmental data of each type of asthma patient, the peak expiratory flow fluctuation performance under each cluster is multiplied by the relative variability fluctuation to obtain the first data; The first data of all clusters corresponding to each type of environmental data of the asthma patient are summed to obtain the second data; The second data is normalized to obtain the asthma response sensitivity of the asthma patients to each type of environmental data.

6. The asthma risk early warning method based on respiratory monitoring according to claim 1, characterized in that, The process of constructing a long-term linked environmental data type set based on the environmental data, and obtaining the influence degree of the multi-linked environmental data patterns of the asthma patients based on the long-term linked environmental data type set, includes: Based on the daily environmental data of the asthma patients, the degree of continuous change of each type of environmental data for each day is obtained; Based on the continuous transformation degree, a long-term continuous transformation curve for each type of environmental data is constructed. Obtain the correlation between the long-term continuous transformation curves of any two environmental data, filter out environmental data pairs with a correlation greater than a preset threshold, and construct the long-term linkage environmental data type set based on the data type of the environmental data pairs; Based on the correlation between the environmental data corresponding to the long-term linked environmental data type set, the influence of the multi-linked environmental data pattern of the asthma patient is obtained.

7. The asthma risk early warning method based on respiratory monitoring according to claim 1, characterized in that, The multi-dimensional environmental risk impact coefficient of the asthma patient is obtained based on the asthma response sensitivity and the impact of the multi-linked environmental data pattern, including: Based on the physiological data, the relative abnormality of the peak expiratory flow rate of the asthma patients was obtained; Based on the influence of the multi-linkage environment data pattern, the anomalous influence of the multi-linkage environment data pattern of the asthma patient is obtained; Based on the asthma response sensitivity, the degree of abnormality in single-type environmental data of the asthma patient is obtained; Based on the relative abnormality of the peak expiratory flow rate, the abnormal impact of the multi-linked environmental data pattern, and the abnormal performance of the single-type environmental data, the multi-dimensional environmental risk impact coefficient of the asthma patient is obtained.

8. The asthma risk early warning method based on respiratory monitoring according to claim 1, characterized in that, The method further includes: Based on the output of the neural network model, an asthma risk warning signal is generated and sent to the asthma patient or the asthma patient's attending physician via a smart device.

9. A method for asthma risk early warning based on respiratory monitoring according to any one of claims 1-8, characterized in that, The neural network model is trained using a training set, which includes multiple training samples from the asthma patients. These training samples include the asthma patients' physiological data, multi-dimensional environmental risk impact coefficients, and corresponding asthma risk warning levels.

10. An asthma risk early warning system based on respiratory monitoring, characterized in that, The system includes a server, and the server includes: A memory on which computer programs are stored; A processor for executing the computer program in the memory to implement the steps of the asthma risk warning method based on respiratory monitoring as described in any one of claims 1-9.

Images