A method of predicting the risk of an allergic disease episode in a child related to weather

Abstract

This invention discloses a method for predicting the risk of weather-related allergic diseases in children, relating to the field of meteorological data processing technology. By integrating and dynamically correcting meteorological conditions and allergen concentrations, the method enables risk prediction to move beyond relying solely on pollen or spore concentrations. It introduces an air diffusion index (Adi) and further forms an air diffusion index set (Adc), which, combined with an allergen parameter set (Allc), dynamically reflects the impact of air dilution conditions on children's actual exposure levels, thereby obtaining a more scientific risk score (Rsc). This method can promptly output a risk level (Lev) and trigger indicator (Trig) that better reflect the real environment when there are significant weather fluctuations, avoiding misleading results caused by single values. It compensates for the shortcomings of existing methods that ignore meteorological dilution and volatility, and makes the risk prediction results real-time, targeted, and operable, providing clear reference information in daily life scenarios.

Description

Technical Field

[0001] This invention relates to the field of meteorological data processing technology, specifically a method for predicting the risk of weather-related allergic diseases in children. Background Technology

[0002] In pediatric health research and clinical observation, the onset of diseases such as allergic asthma and allergic rhinitis is closely related to weather conditions. With rapid urbanization, the coupling effect of air pollution and meteorological environment is becoming increasingly complex, leading to a continuous rise in the frequency and intensity of allergy attacks in children. How to establish a more accurate attack risk prediction mechanism by incorporating weather factors has become an urgent problem to be solved.

[0003] In existing allergy risk prediction models, researchers typically rely solely on parameters such as temperature, relative humidity, and pollen concentration for trend-based risk assessments. However, atmospheric diffusion capacity—the degree to which pollutants and allergens are diluted in space—is rarely included as a predictor. In reality, even if temperature and humidity remain within normal ranges, insufficient atmospheric diffusion capacity can lead to a significant accumulation of allergens near the ground, resulting in high-risk exposure. The shortcoming of existing methods lies in neglecting changes in atmospheric dilution capacity, causing some predictions to be distorted, leading to either underestimation or overestimation of risk. Summary of the Invention

[0004] To address the shortcomings of existing technologies, this invention provides a method for predicting the risk of weather-related allergic diseases in children, thus solving the problems mentioned in the background section.

[0005] To achieve the above objectives, the present invention provides the following technical solution: a method for predicting the risk of weather-related allergic diseases in children, comprising the following steps: S1. Collect meteorological parameter sets through the meteorological data interface, organize them according to the time axis, and form the raw data set; S2. Based on wind speed at each time point, calculate the air diffusion index at each time point, and obtain the air diffusion index set after summarizing all time points. S3. Using pollen concentration and spore concentration provided by pollen monitoring points, an allergen parameter set is formed. The allergen parameter set is then fused with the air diffusion index set Adc to obtain a risk score. S4. Based on the fluctuation range of the air diffusion index set at different time points and the superposition effect of the allergen parameter set, the risk score is corrected and evaluated, and the risk level and trigger indication are output. S5. Based on the risk level and trigger indication, generate risk assessment, prediction, and early warning and push them out.

[0006] Preferably, S1 includes S11; S11. Download wind speed Wsp data in real time through meteorological monitoring stations and regional meteorological service interfaces deployed in the monitoring area, and collect data at hourly or minutely sampling intervals. Atmospheric boundary layer temperature profiles are obtained using radiosondes, and the mixing layer height Mlh is calculated using the abrupt change points in the temperature lapse rate within the profiles. Wind speeds Wsp and mixed layer heights Mlh acquired at different time points are paired according to timestamps to form raw meteorological data.

[0007] Preferably, S2 includes S21; S21. In the raw meteorological data, the wind speed Wsp and the mixing layer height Mlh at each time point are used as inputs. The minimum-maximum normalization method is used to normalize the data to eliminate the differences between the dimensions. The air diffusion index Adi is then calculated for the normalized wind speed Wsp and the mixing layer height Mlh at the corresponding time points. The air diffusion index Adi is obtained using the following formula: ; In the formula, This represents the fluctuation penalty coefficient, with a value range of [0, 1], used to adjust the degree to which fluctuations weaken the diffusion capability; and These represent the normalized volatility of wind speed and the normalized volatility of the mixing layer height, respectively. Among them, the normalized volatility of wind speed Defined as the average absolute difference between normalized wind speeds Wsp at adjacent time points; normalized volatility of the mixing layer height. It is defined as the average absolute difference of the hybrid layer height Mlh after normalization between adjacent time points.

[0008] Preferably, S2 further includes S22; S22. After completing the hourly single-value calculation of the air diffusion index Adi, collect the air diffusion index Adi at different time points in chronological order, perform a consistency check, and construct the air diffusion index set Adc. Among them, the consistency check performs logical verification on the air diffusion index Adi in the sequence based on the threshold judgment method, and removes erroneous values ​​caused by missing or abnormal input data; and supplements the missing calculation points after removing erroneous values, specifically by using the nearest time interpolation method.

[0009] Preferably, S3 includes S31; S31. By setting up air particle monitoring points in the target area and using a volumetric air sampler, particulate matter in the air is continuously collected. Then, an automatic optical image recognition system is used to identify and count the number of pollen grains per unit volume of air, and the pollen concentration Pol is calculated. An air fungus sampler was used to collect air samples; The collected air samples were observed under a microscope to directly count the spores, thereby obtaining the number of fungal spores per unit volume of air and thus the spore concentration Spo. The pollen concentration Pol, spore concentration Spo, and air diffusion index set Adc are time-stamped and then integrated to obtain the feature set Fea.

[0010] Preferably, S3 further includes S32; S32. Based on the obtained feature set Fea, the pollen concentration Pol and spore concentration Spo are normalized using the minimum-maximum normalization method to eliminate the dimensional differences between different data, and then replaced in the feature set Fea. Perform risk assessment on feature set Fea and obtain risk score Rsc; The risk score Rsc is obtained through Obtain the calculation formula.

[0011] Preferably, S4 includes S41; S41. Based on the obtained risk score Rsc, the dynamic fluctuation characteristics of the air diffusion index set Adc are introduced for secondary correction: Calculate the mean and standard deviation of the air diffusion index set Adc within a preset time window, and label them as the mean Amu and standard deviation Asd; Then, the correction factor Adj is calculated based on the mean Amu and the standard deviation Asd; The correction factor Adj is obtained by the formula Adj = (1-Amu) × Asd; The risk score Rsc is then corrected a second time based on the correction factor Adj to obtain the corrected risk score Rscm. The revised risk score Rscm is obtained by the formula Rscm=Rsc×(1+Adj); In the calculation process, the modified risk score Rscm performs logical verification on the modification factor Adj based on the range constraint method: when the modification factor Adj < 0, it is truncated to 0, and when the modification factor Adj > 1, it is truncated to 1.

[0012] Preferably, S4 further includes S42; S42. Based on the threshold grading method, set the boundary for the obtained corrected risk score Rscm, and obtain the risk level Lev and trigger indication Trig. When the corrected risk score Rscm∈[0,30], the risk level Lev is set to low, and no trigger indication Trig is generated. When the corrected risk score Rscm ∈ [31, 70], set the risk level Lev = medium state and generate a trigger indicator Trig. When the corrected risk score Rscm∈[71,100], the risk level Lev is set to high state, and a trigger indicator Trig is generated.

[0013] Preferably, S5 includes S51; S51. Based on the risk level Lev and the trigger indicator Trig, initiate the warning content generation process; use a template-based assembly method to match warning templates according to different risk levels Lev, and combine real-time data: risk score Rscm, pollen concentration Pol, spore concentration Spo and air diffusion index Adi to fill the warning template content and generate warning content Msg.

[0014] Preferably, S5 further includes S52; S52. When the trigger indicator Trig is activated, the warning content Msg will be sent to the user terminal through multiple channels: mobile terminal, email, dedicated information platforms of schools and medical institutions, public web page announcements and social media interfaces; After the warning content Msg is pushed, the warning content Msg, the corresponding risk level Lev, the corrected risk score Rscm, and the trigger indicator Trig are stored together as a warning record Rec for archiving and subsequent analysis. When any user-side channel fails to push the alert, the same alert message (Msg) will be resent to the user-side channel's backup channel via a resending mechanism.

[0015] This invention provides a method for predicting the risk of weather-related allergic disease attacks in children, which has the following beneficial effects: (1) By combining the allergen parameter set Allc to dynamically reflect the impact of air dilution conditions on children's actual exposure levels, a more scientific risk score Rsc can be obtained. Compared with existing methods, it can output a risk level Lev and trigger indicator Trig that are more in line with the real environment in a timely manner through a correction mechanism when there are large weather fluctuations, avoiding the misleading effect of a single value. It makes up for the shortcomings of existing methods that ignore weather dilution and fluctuations, and also makes the risk prediction results real-time, targeted and operable, providing parents with clear and scientific reference in daily life scenarios.

[0016] (2) After normalizing the pollen concentration (Pol) and spore concentration (Spo), the calculation bias caused by differences in measurement units and concentration ranges among different allergens was further eliminated, so that the final risk score (Rsc) can truly reflect the combined effect of the total intensity of allergens and air dilution capacity. Compared with the traditional method that relies solely on pollen index or a single concentration value as the basis for risk assessment, this method can simultaneously consider the cumulative effects of multiple allergens on children's health and correct for meteorological dilution conditions. This allows parents and schools to take protective measures earlier in the actual environment, reducing the probability of sudden onset of allergic rhinitis or asthma in children.

[0017] (3) A secondary correction based on the fluctuation characteristics of the air diffusion index set Adc was performed using the risk score Rsc to obtain a corrected risk score Rscm that better reflects the actual situation. Furthermore, a grading method can be used to output clear risk levels Lev and corresponding trigger indicators Trig, achieving a closed-loop process from numerical values ​​to grading and then to action. When the trigger indicator Trig is activated, a structured warning content Msg is automatically generated. This content not only includes summary information on core parameters such as pollen concentration Pol, spore concentration Spo, and air diffusion index Adi, but also provides trend indicators and specific protective measures, enabling parents and schools to immediately understand the level of risk and take action. Furthermore, a multi-channel push mechanism ensures that the warning content Msg is delivered to the user terminal in the first instance. Even if a push fails on one channel, it can still be resent through a backup channel, avoiding information loss. Attached Figure Description

[0018] Figure 1 This is a schematic diagram illustrating the steps of a method for predicting the risk of weather-related allergic diseases in children according to the present invention. Figure 2 This is a schematic diagram comparing the risk score Rsc and the corrected risk score Rscm. Detailed Implementation

[0019] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.

[0020] Example 1: This invention provides a method for predicting the risk of weather-related allergic disease attacks in children. Please refer to [link / reference]. Figure 1 This includes the following steps: S1. Collect meteorological parameter sets through the meteorological data interface, organize them according to the time axis, and form the raw data set; S2. Based on the wind speed Wsp at each time point, calculate the air diffusion index Adi at each time point, and after summing all time points, obtain the air diffusion index set Adc. S3. The pollen concentration Pol and spore concentration Spo provided by the pollen monitoring points are used to form the allergen parameter set Allc. The allergen parameter set Allc is fused with the air diffusion index set Adc to obtain the risk score Rsc. S4. Based on the superposition effect of the fluctuation range of the air diffusion index set Adc at different time points and the allergen parameter set Allc, the risk score Rsc is corrected and evaluated, and the risk level Lev and trigger indicator Trig are output. S5. Based on the risk level (Lev) and trigger indicator (Trig), generate risk assessment predictions and early warnings and push them out.

[0021] In this embodiment, steps S1 to S5 achieve multi-dimensional fusion and dynamic correction of meteorological conditions and allergen concentrations. This allows risk prediction to move beyond relying solely on pollen or spore concentrations, incorporating the air diffusion index Adi and further forming an air diffusion index set Adc. Combined with the allergen parameter set Allc, this dynamically reflects the impact of air dilution conditions on children's actual exposure levels, resulting in a more scientific risk score Rsc. Compared to existing methods, this approach can promptly output a risk level Lev and trigger indication Trig that better reflect the real environment when meteorological fluctuations are significant, avoiding misleading results from single values. For example, if a high pollen concentration Pol is detected, and the air diffusion index Adi indicates strong atmospheric dilution, the risk score Rsc can be reasonably lowered, placing the risk level Lev at a medium-risk level. This prevents parents from excessively restricting children's outdoor activities due to misjudgment. Conversely, when the spore concentration Spo is at a medium level but the air diffusion index set Adc shows consistently low values ​​and significant fluctuations, a Trig indicator is triggered, generating a high-risk warning to remind parents to reduce children's outdoor activities and use masks, effectively reducing the risk of allergic asthma or allergic rhinitis attacks. This approach not only compensates for the shortcomings of existing methods that ignore meteorological dilution and volatility, but also makes the risk prediction results real-time, targeted, and operable, providing parents with clear and scientific references in daily life scenarios.

[0022] Example 2: Specifically: S1 includes S11; S11. Download wind speed Wsp data in real time through meteorological monitoring stations and regional meteorological service interfaces deployed in the monitoring area, and collect data at hourly or minutely sampling intervals. Atmospheric boundary layer temperature profiles are obtained using radiosondes, and the mixing layer height Mlh is calculated using the abrupt change points in the temperature lapse rate within the profiles. The wind speed Wsp and the mixing layer height Mlh obtained at different time points are paired according to the timestamp to form the raw meteorological data Raw. It should be noted that: Mixing layer height Mlh: refers to the height at which air in the atmospheric boundary layer can be fully mixed through turbulent motion; it is also known as the atmospheric mixing layer height. Below this altitude, gases and particulate matter in the air mix more evenly in the vertical direction; above this altitude, the atmosphere is more stable, and pollutants and particulate matter are less likely to rise and diffuse. The mixing layer height Mlh directly determines the vertical dilution space of pollutants and allergens in the atmosphere; When the mixed layer height Mlh is low, pollutants and allergens are confined to the near-surface layer, leading to increased concentrations and increasing the risk of allergic diseases in children. When the mixing layer height Mlh is high, particulate matter in the air can be diluted into a larger space, the concentration near the ground layer is reduced, and the risk is correspondingly reduced. Therefore, combining the mixing layer height Mlh with the wind speed Wsp to calculate the air diffusion index Adi can more accurately reflect the air dilution and diffusion capacity. The mixing layer height Mlh reflects the spatial range in which the atmosphere can dilute pollutants, particulate matter, and allergens upwards; it does not directly measure the height of a child's breathing zone, but rather is an atmospheric dilution spatial parameter that controls the background concentration of a child's breathing zone. To give a direct comparison: If the mixing layer height Mlh is only 100m, it means that the space for air dilution is limited, allergens accumulate in the near-surface layer, and the concentration in the children's breathing layer will be higher; If the mixed layer height Mlh is 1500m, allergens can be diluted over a larger space, and the concentration in the near-surface layer (1–2m) will be significantly reduced.

[0023] S2 includes S21; S21. In the raw meteorological data, the wind speed Wsp and the mixing layer height Mlh at each time point are used as inputs. The minimum-maximum normalization method is used to normalize the data to eliminate the differences between the dimensions. The air diffusion index Adi is then calculated for the normalized wind speed Wsp and the mixing layer height Mlh at the corresponding time points. The air diffusion index Adi is obtained using the following formula: ; In the formula, This represents the fluctuation penalty coefficient, with a value range of [0, 1], used to adjust the degree to which fluctuations weaken the diffusion capability; and These represent the normalized volatility of wind speed and the normalized volatility of the mixing layer height, respectively. Among them, the normalized volatility of wind speed Defined as the average absolute difference between normalized wind speeds Wsp at adjacent time points; normalized volatility of the mixing layer height. It is defined as the average absolute difference of the hybrid layer height Mlh after normalization between adjacent time points.

[0024] S2 further includes S22; S22. After completing the hourly single-value calculation of the air diffusion index Adi, collect the air diffusion index Adi at different time points in chronological order, perform a consistency check, and construct the air diffusion index set Adc. Among them, the consistency check performs logical verification on the air diffusion index Adi in the sequence based on the threshold judgment method, and removes erroneous values ​​caused by missing or abnormal input data; and supplements the missing calculation points after removing erroneous values, specifically by using the nearest time interpolation method; It should be noted that: Threshold determination method: Anomalies are identified by setting a reasonable range for the air diffusion index Adi (e.g., 0 to 1) to ensure the physical rationality of the numerical values; The air diffusion index set Adc is composed of air diffusion indices Adi arranged in chronological order after logical verification and correction, and is used to reflect the dynamic characteristics of air diffusion within a certain time range.

[0025] In this embodiment, the processing steps S1 to S2 eliminate the dimensional differences between wind speed Wsp and mixing layer height Mlh during the acquisition and calculation of raw meteorological data (Raw). By introducing normalization and volatility correction, the calculated air diffusion index Adi not only reflects the instantaneous dilution capacity of the atmosphere but also exhibits sensitive response to short-term fluctuations in meteorological conditions. Furthermore, logical verification and interpolation of the air diffusion index Adi at each time point result in an air diffusion index set Adc that possesses continuity and physical rationality, enabling dynamic tracking of changes in air diffusion capacity. Compared to traditional methods that rely solely on wind speed at a single moment or local weather conditions to assess risk, this new method offers several alternatives. For example, if the mixing layer height (Mlh) remains high for several hours, the air diffusion index (Adc) can remain at a moderate level, thus avoiding misjudging high risk based solely on low wind speed. Conversely, when the mixing layer height (Mlh) remains low for an extended period and the wind speed (Wsp) is also low, the air diffusion index (Adc) will drop rapidly, indicating a trend of allergen accumulation in the near-surface layer. This makes the prediction results more consistent with the actual environment, helping parents take preventative measures in advance during periods of calm and stable weather.

[0026] Example 3: Specifically: S3 includes S31; S31. By deploying air particle monitoring points in the target area and using volumetric air samplers, such as Hirst pollen samplers, to continuously collect particulate matter in the air; Then, an automatic optical image recognition system is used to identify and count the number of pollen grains per unit volume of air, and the pollen concentration Pol is calculated. Air samples are collected using air fungus samplers, such as impact samplers and filter samplers; The collected air samples were observed under a microscope to directly count the spores, thereby obtaining the number of fungal spores per unit volume of air and thus the spore concentration Spo. The pollen concentration Pol and spore concentration Spo are time-stamped with the air diffusion index set Adc, and then integrated to obtain the feature set Fea. It should be noted that: Pollen concentration Pol: refers to the concentration of air pollen obtained by collecting air samples through a conventional air particle sampler and counting them manually or automatically. The unit is the number of particles per m³. Spore concentration Spo: refers to the concentration of fungal spores obtained by collecting air samples using an air fungus sampler and counting them under a microscope or by culture, and is expressed in particles per m³. Feature set Fea: A fusion set consisting of air diffusion index Adi, pollen concentration Pol, and spore concentration Spo, which is the input dataset for risk score calculation.

[0027] S3 further includes S32; S32. Based on the obtained feature set Fea, the pollen concentration Pol and spore concentration Spo are normalized using the minimum-maximum normalization method to eliminate the dimensional differences between different data, and then replaced in the feature set Fea. Perform risk assessment on feature set Fea and obtain risk score Rsc; The risk score Rsc is obtained through Obtain the calculation formula; It should be noted that: The risk score Rsc is a quantitative indicator calculated based on allergen concentration and airborne diffusion capacity, used to measure the risk of allergic disease onset in children.

[0028] In this embodiment, the processing steps described in step S3 align pollen concentration (Pol) and spore concentration (Spo) with the air diffusion index set (Adc) over time, integrating them into a unified feature set (Fea). This ensures the data relied upon for risk calculation remains consistent in both time and physical dimensions. After normalizing pollen concentration (Pol) and spore concentration (Spo), calculation biases caused by differences in measurement units and concentration ranges among different allergens are further eliminated. This ensures that the final risk score (Rsc) accurately reflects the combined effect of the total allergen intensity and air dilution capacity. Compared to traditional methods that rely solely on pollen index or a single concentration value for risk assessment, this approach simultaneously considers the cumulative effects of multiple allergens on children's health and incorporates corrections based on meteorological dilution conditions. For example, during the peak pollen season in spring, if the pollen concentration Pol is high but the spore concentration Spo is low, and the air diffusion index set Adc remains at a moderate value, the calculated risk score Rsc will be at a moderate level, avoiding the misjudgment of high risk due to a single high pollen concentration. Conversely, in autumn, if both the pollen concentration Pol and the spore concentration Spo increase simultaneously, and the air diffusion index set Adc remains at a persistently low value, the risk score Rsc will increase significantly. This allows parents and schools to take protective measures earlier in the actual environment, reducing the probability of sudden onset of allergic rhinitis or asthma in children.

[0029] Example 4: Please see Figure 1 and Figure 2 Specifically: S4 includes S41; S41. Based on the obtained risk score Rsc, the dynamic fluctuation characteristics of the air diffusion index set Adc are introduced for secondary correction: Calculate the mean and standard deviation of the air diffusion index set Adc within a preset time window (e.g., 3 hours or 6 hours), denoted as mean Amu and standard deviation Asd; Then, the correction factor Adj is calculated based on the mean Amu and the standard deviation Asd; The correction factor Adj is obtained by the formula Adj=(1-Amu)×Asd; the physical meaning of this formula is that when the mean Amu is low (poor overall diffusion) or the standard deviation Asd is high (large fluctuations, instability), the correction factor Adj increases; conversely, it decreases. The risk score Rsc is then corrected a second time based on the correction factor Adj to obtain the corrected risk score Rscm. The revised risk score Rscm is obtained by the formula Rscm=Rsc×(1+Adj); In the calculation process, the modified risk score Rscm performs logical verification on the modification factor Adj based on the range constraint method: when the modification factor Adj < 0, it is truncated to 0, and when the modification factor Adj > 1, it is truncated to 1.

[0030] S4 also includes S42; S42. Based on the threshold grading method, set the boundary for the obtained corrected risk score Rscm, and obtain the risk level Lev and trigger indication Trig. When the corrected risk score Rscm∈[0,30], the risk level Lev is set to low, and no trigger indication Trig is generated. When the corrected risk score Rscm ∈ [31, 70], set the risk level Lev = medium state and generate a trigger indicator Trig. When the corrected risk score Rscm ∈ [71, 100], set the risk level Lev = high state and generate a trigger indicator Trig. It should be noted that: Risk level Lev represents the state quantity of the prediction result. It is a static description, that is, the current level of risk. When the risk level Lev is determined to be "medium risk" or "high risk", a trigger indicator Trig is automatically generated. Trigger indicator Trig serves as a control signal, automatically generated when the risk level Lev reaches the threshold. It is used to activate the warning content generation module and the warning distribution module, ensuring that the warning generation and distribution process is conditionally triggered and automatically executed. At the same time, Trigger indicator Trig is recorded in the warning record Rec for subsequent traceability and effect evaluation.

[0031] S5 includes S51; S51. Based on the risk level Lev and the trigger indicator Trig, start the warning content generation process; use a template assembly method to match the warning template according to different risk levels Lev, and combine real-time data: risk score Rscm, pollen concentration Pol, spore concentration Spo and air diffusion index Adi to fill the warning template content and generate the warning content Msg. The warning message Msg includes the following specific parts: Risk level description: Clearly display the risk level (Lev), such as "High Risk"; Monitoring Parameter Summary: Real-time display of key early warning values ​​related to pollen concentration (Pol), spore concentration (Spo), and air diffusion index (Adi); Trend indication: Based on the changing trend of the revised risk score Rscm, the risk may continue, increase, or decrease. Early warning and protection measures include: Reduce outdoor activity time and avoid going out during high-risk periods; If you must go out, it is recommended to wear a protective mask (such as an N95 or a children's protective mask). Use air purification equipment indoors and maintain good ventilation; Strengthen personal hygiene, such as washing your face and hands immediately after returning from outside to avoid pollen or spore residue; For children diagnosed with allergies or asthma, it is recommended to carry emergency medication prescribed by a doctor with them at all times.

[0032] S5 also includes S52; S52. When the trigger indicator Trig is activated, the warning content Msg will be sent to the user terminal through multiple channels: mobile terminal (such as APP notification or SMS), email, dedicated information platforms of schools and medical institutions, public web page announcements and social media interfaces; After the warning content Msg is pushed, the warning content Msg, the corresponding risk level Lev, the corrected risk score Rscm, and the trigger indicator Trig are stored together as a warning record Rec for archiving and subsequent analysis. When any user-side channel fails to push the alert, the same alert message (Msg) will be resent to the user-side channel's backup channel through a resending mechanism. It should be noted that: Multi-channel push method: refers to using multiple communication channels to send early warning information simultaneously in order to improve the delivery rate and timeliness; Warning Log (Rec): Records the core data of each warning issued (content Msg, level Lev, score Rscm, trigger Trig) for easy traceability and effect evaluation; Resend Mechanism: When a push method fails, a backup channel is automatically activated to resend the message, ensuring information coverage.

[0033] In this embodiment, steps S4 and S5 not only perform a secondary correction on the initially obtained risk score Rsc based on the fluctuation characteristics of the air diffusion index set Adc, resulting in a more accurate and realistic risk score Rscm, but also utilize a grading method to output a clear risk level Lev and corresponding trigger indicator Trig, achieving a closed-loop process from numerical risk assessment to grading and action. When the trigger indicator Trig is activated, a structured warning message Msg is automatically generated. This message includes summary information on core parameters such as pollen concentration Pol, spore concentration Spo, and air diffusion index Adi, as well as trend indicators and specific protective measures, enabling parents and schools to immediately understand the risk level and take action. Furthermore, a multi-channel push mechanism ensures that the warning message Msg is delivered to the user terminal in a timely manner. Even if a push fails on one channel (e.g., due to mobile terminal network failure), it can still be resent through a backup channel, preventing information loss. For example, during periods of stable weather in spring, when the revised risk score Rscm is assessed as high-risk, schools can immediately receive a warning message (Msg) instructing them to "reduce outdoor activities and wear children's protective masks." Parents will also receive the same message on their mobile phones, ensuring comprehensive protection for children in high-risk environments. Simultaneously, all issued warning messages (Msg), along with their corresponding risk levels (Lev), revised risk scores (Rscm), and trigger indicators (Trig), are recorded in the warning log (Rec), providing comprehensive data support for subsequent effectiveness evaluation and follow-up.

[0034] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and variations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A method for predicting the risk of weather-related allergic disease attacks in children, characterized in that: Includes the following steps: S1. Collect meteorological parameter sets through the meteorological data interface, organize them according to the time axis, and form the raw data set; S2. Based on wind speed at each time point, calculate the air diffusion index at each time point, and obtain the air diffusion index set after summarizing all time points. S3. Using pollen concentration and spore concentration provided by pollen monitoring points, an allergen parameter set is formed. The allergen parameter set is then fused with the air diffusion index set Adc to obtain a risk score. S4. Based on the fluctuation range of the air diffusion index set at different time points and the superposition effect of the allergen parameter set, the risk score is corrected and evaluated, and the risk level and trigger indication are output. S5. Based on the risk level and trigger indication, generate risk assessment, prediction, and early warning and push them out.

2. The method for predicting the risk of weather-related allergic diseases in children according to claim 1, characterized in that: S1 includes S11; S11. Download wind speed Wsp data in real time through meteorological monitoring stations and regional meteorological service interfaces deployed in the monitoring area, and collect data at hourly or minutely sampling intervals. Atmospheric boundary layer temperature profiles are obtained using radiosondes, and the mixing layer height Mlh is calculated using the abrupt change points in the temperature lapse rate within the profiles. Wind speeds Wsp and mixed layer heights Mlh acquired at different time points are paired according to timestamps to form raw meteorological data.

3. The method for predicting the risk of weather-related allergic diseases in children according to claim 2, characterized in that: S2 includes S21; S21. In the raw meteorological data, the wind speed Wsp and the mixing layer height Mlh at each time point are used as inputs. The minimum-maximum normalization method is used to normalize the data to eliminate the differences between the dimensions. The air diffusion index Adi is then calculated for the normalized wind speed Wsp and the mixing layer height Mlh at the corresponding time points. The air diffusion index Adi is obtained using the following formula: ； In the formula, This represents the fluctuation penalty coefficient, with a value range of [0, 1]. and These represent the normalized volatility of wind speed and the normalized volatility of the mixing layer height, respectively. Among them, the normalized volatility of wind speed Defined as the average absolute difference between normalized wind speeds Wsp at adjacent time points; normalized volatility of the mixing layer height. It is defined as the average absolute difference of the hybrid layer height Mlh after normalization between adjacent time points.

4. The method for predicting the risk of weather-related allergic diseases in children according to claim 3, characterized in that: S2 further includes S22; S22. After completing the hourly single-value calculation of the air diffusion index Adi, collect the air diffusion index Adi at different time points in chronological order, perform a consistency check, and construct the air diffusion index set Adc. Among them, the consistency check performs logical verification on the air diffusion index Adi in the sequence based on the threshold judgment method, and removes erroneous values ​​caused by missing or abnormal input data; and supplements the missing calculation points after removing erroneous values, specifically by using the nearest time interpolation method.

5. The method for predicting the risk of weather-related allergic diseases in children according to claim 4, characterized in that: S3 includes S31; S31. By setting up air particle monitoring points in the target area and using a volumetric air sampler, particulate matter in the air is continuously collected. Then, an automatic optical image recognition system is used to identify and count the number of pollen grains per unit volume of air, and the pollen concentration Pol is calculated. An air fungus sampler was used to collect air samples; The collected air samples were observed under a microscope to directly count the spores, thereby obtaining the number of fungal spores per unit volume of air and thus the spore concentration Spo. The pollen concentration Pol, spore concentration Spo, and air diffusion index set Adc are time-stamped and then integrated to obtain the feature set Fea.

6. The method for predicting the risk of weather-related allergic disease attacks in children according to claim 5, characterized in that: S3 further includes S32; S32. Based on the obtained feature set Fea, the pollen concentration Pol and spore concentration Spo are normalized using the minimum-maximum normalization method to eliminate the dimensional differences between different data, and then replaced in the feature set Fea. Perform risk assessment on feature set Fea and obtain risk score Rsc; The risk score Rsc is obtained through Obtain the calculation formula.

7. The method for predicting the risk of weather-related allergic disease attacks in children according to claim 6, characterized in that: S4 includes S41; S41. Based on the obtained risk score Rsc, the dynamic fluctuation characteristics of the air diffusion index set Adc are introduced for secondary correction: Calculate the mean and standard deviation of the air diffusion index set Adc within a preset time window, and label them as the mean Amu and standard deviation Asd; Then, the correction factor Adj is calculated based on the mean Amu and the standard deviation Asd; The correction factor Adj is obtained by the formula Adj = (1-Amu) × Asd; The risk score Rsc is then corrected a second time based on the correction factor Adj to obtain the corrected risk score Rscm. The revised risk score Rscm is obtained by the formula Rscm=Rsc×(1+Adj); In the calculation process, the modified risk score Rscm performs logical verification on the modification factor Adj based on the range constraint method: when the modification factor Adj < 0, it is truncated to 0, and when the modification factor Adj > 1, it is truncated to 1.

8. The method for predicting the risk of weather-related allergic diseases in children according to claim 7, characterized in that: S4 also includes S42; S42. Based on the threshold grading method, set the boundary for the obtained corrected risk score Rscm, and obtain the risk level Lev and trigger indication Trig. When the corrected risk score Rscm∈[0,30], the risk level Lev is set to low, and no trigger indication Trig is generated. When the corrected risk score Rscm ∈ [31, 70], set the risk level Lev = medium state and generate a trigger indicator Trig. When the corrected risk score Rscm∈[71,100], the risk level Lev is set to high state, and a trigger indicator Trig is generated.

9. The method for predicting the risk of weather-related allergic diseases in children according to claim 8, characterized in that: S5 includes S51; S51. Based on the risk level Lev and the trigger indicator Trig, initiate the warning content generation process; Using a template-based assembly method, warning templates are matched according to different risk levels (Lev), and real-time data (risk score Rscm, pollen concentration Pol, spore concentration Spo, and air diffusion index Adi) are combined to fill the warning template content and generate warning content Msg.

10. A method for predicting the risk of weather-related allergic disease attacks in children according to claim 9, characterized in that: S5 also includes S52; S52. When the trigger indicator Trig is activated, the warning content Msg will be sent to the user terminal through multiple channels: mobile terminal, email, dedicated information platforms of schools and medical institutions, public web page announcements and social media interfaces; After the warning content Msg is pushed, the warning content Msg, the corresponding risk level Lev, the corrected risk score Rscm, and the trigger indicator Trig are stored together as a warning record Rec for archiving and subsequent analysis. When any user-side channel fails to push the alert, the same alert message (Msg) will be resent to the user-side channel's backup channel via a resending mechanism.

Images