A method for assessing and predicting the risk of childhood asthma attacks

Abstract

The present invention discloses a method for assessing and predicting the risk of childhood asthma attacks, which relates to the technical field of medical data processing. The method includes collecting the physiological data and environmental humidity data of children; performing fluctuation analysis on the respiratory rate data to obtain a respiratory fluctuation index, and at the same time performing correlation analysis on the heart rate data and body temperature data to obtain body state parameters, and performing cross-calculation to generate an initial asthma risk score; retrieving a reference data group matching the age of the child from a preset historical database, comparing and analyzing the initial asthma risk score with the reference data group to obtain a correction coefficient; multiplying the initial asthma risk score by the correction coefficient to generate a final risk prediction value, and determining the asthma attack risk level according to the numerical range of the final risk prediction value. By collecting children's physiological data and environmental humidity data and combining data processing and analysis methods, the present invention can monitor the physical condition and environmental factors of children in real time and assess the risk of asthma attacks.

Description

Technical Field

[0001] The present invention relates to the technical field of medical data processing, and in particular to a method for assessing and predicting the risk of childhood asthma attacks. Background Art

[0002] Childhood asthma, as a common chronic respiratory disease, has been showing a continuous upward trend in the global incidence rate. Traditional asthma management mainly relies on clinical symptom observation and pulmonary function testing. This passive monitoring method often fails to achieve early warning of asthma attacks. With the rapid development of Internet of Things technology and biosensors, intelligent health monitoring systems based on multi-source heterogeneous data have gradually become a research hotspot. Existing childhood asthma monitoring technologies mainly adopt single physiological parameter monitoring or simple multi-parameter superposition analysis. These methods still have obvious deficiencies in aspects such as the continuity of data collection, the fusion analysis of multi-dimensional data, and individualized assessment. Especially in key technical aspects such as the coupled analysis of environmental factors and physiological indicators, the real-time nature of dynamic risk assessment, and the adaptability of prediction models, no systematic solutions have been formed.

[0003] The main technical difficulties in the current field of childhood asthma monitoring include: traditional single-parameter monitoring methods cannot comprehensively reflect the respiratory status of children, and the data collection process often has discontinuity, affecting the accuracy of assessment; secondly, existing multi-parameter fusion algorithms generally adopt linear superposition models, failing to fully consider the non-linear correlation effects and time-series characteristics between parameters; thirdly, the influence mechanism of environmental factors on asthma attacks is complex, and existing technologies rarely consider the dynamic coupling relationship between environmental parameters and physiological indicators; finally, the physiological characteristics of children in different age groups vary significantly, lacking targeted individualized assessment criteria and adaptive prediction models. These technical bottlenecks severely restrict the clinical application effect of the asthma early warning system. Summary of the Invention

[0004] In view of the problems existing in the prior art, the present invention proposes a method for assessing and predicting the risk of childhood asthma attacks, which can solve the problems mentioned in the background art.

[0005] The technical solution is as follows:

[0006] In the first aspect, the present invention provides a method for assessing and predicting the risk of childhood asthma attacks, which includes,

[0007] Collecting the respiratory rate data, heart rate data, body temperature data and environmental humidity data of children, and inputting the respiratory rate data, the heart rate data, the body temperature data and the environmental humidity data into a preset data processing module;

[0008] Based on the data processing module, perform fluctuation analysis on the respiratory rate data to obtain a respiratory fluctuation index. At the same time, perform correlation analysis on the heart rate data and the body temperature data to obtain body state parameters;

[0009] Perform multi-dimensional cross-calculation on the respiratory fluctuation index, the body state parameters, and the environmental humidity data to generate an initial asthma risk score;

[0010] Retrieve a reference data set matching the age of the child from a preset historical database, and perform comparative analysis on the initial asthma risk score and the reference data set to obtain a correction coefficient;

[0011] Multiply the initial asthma risk score by the correction coefficient to generate a final risk prediction value, and determine the asthma attack risk level according to the numerical range of the final risk prediction value.

[0012] As a preferred solution of the method for assessing and predicting the risk of childhood asthma attack according to the present invention, wherein: obtaining the body state parameters includes:

[0013] Adopt a sliding window method to segment the respiratory rate data, calculate the standard deviation and mean of the respiratory rate data in each sliding window, and calculate the respiratory volatility;

[0014] Cumulatively sum the respiratory volatilities within 24 consecutive hours to obtain a respiratory fluctuation index;

[0015] At the same time, perform correlation analysis on the heart rate data and the body temperature data to obtain a heart rate standard score;

[0016] Convert the body temperature data into a body temperature change curve and calculate the temperature difference value;

[0017] Perform state evaluation on the heart rate standard score and the temperature difference value to obtain the body state parameters.

[0018] As a preferred solution of the method for assessing and predicting the risk of childhood asthma attack according to the present invention, wherein: the respiratory volatility obtains a basic fluctuation value based on the discrete degree of the respiratory signal, and then introduces the change rate of the respiratory state as a regulation factor; when it is detected that the respiratory fluctuation index of the respiratory pattern exceeds a preset threshold, it indicates that the stability of the respiratory system decreases. At this time, the basic fluctuation value is amplified by a regulation coefficient to finally obtain the respiratory volatility;

[0019] The calculation of the respiratory fluctuation index includes:

[0020] Obtain the respiratory volatility for 24 consecutive hours; construct a periodic adjustment function based on the characteristics of the human circadian rhythm; use the periodic adjustment function to weight the respiratory volatility at different time periods; accumulate and sum the weighted respiratory volatility to obtain the respiratory fluctuation index;

[0021] The calculation of the standard heart rate score includes:

[0022] Based on the deviation level of the measured heart rate from the basal heart rate, and standardize this deviation level through the age coefficient. On this basis, use the heart rate change rate as a dynamic adjustment factor, and when a rapid heart rate change is detected, the standard score will be enhanced or inhibited accordingly through the dynamic adjustment coefficient.

[0023] As a preferred embodiment of the method for assessing and predicting the risk of childhood asthma attack according to the present invention, wherein: the generation of the initial asthma risk score includes:

[0024] Extract the time series characteristics of the environmental humidity data and construct a humidity influence model;

[0025] Calculate the dynamic sensitivity index based on the humidity influence model; the dynamic sensitivity index reflects the degree of irritation of the environmental humidity change to the respiratory tract of children;

[0026] Perform a non-linear mapping on the respiratory fluctuation index and construct a dynamic weight function of the body state parameters;

[0027] Calculate the initial asthma risk score based on the above parameter combination and perform dynamic calibration.

[0028] As a preferred embodiment of the method for assessing and predicting the risk of childhood asthma attack according to the present invention, wherein:

[0029] The construction method of the humidity influence model includes:

[0030] Set a reference humidity value as the basic level; construct a plurality of Gaussian functions with characteristic time points, fluctuation amplitudes and time scale parameters; superimpose the reference humidity value with the plurality of Gaussian functions to obtain the humidity influence model;

[0031] The dynamic sensitivity index The calculation of includes:

[0032] Obtain the first derivative of the humidity influence model with respect to time to obtain the humidity change rate;

[0033] Obtain the second derivative of the humidity influence model with respect to time to obtain the humidity change acceleration;

[0034] Multiply the humidity change acceleration by a preset acceleration influence factor and then add 1 to obtain a dynamic adjustment coefficient;

[0035] Multiply the absolute value of the humidity change rate by the dynamic adjustment coefficient and perform an integration operation within the observation time period;

[0036] Divide the integration result by the observation time period to obtain the dynamic sensitivity index.

[0037] As a preferred embodiment of the method for assessing and predicting the risk of childhood asthma attacks according to the present invention, wherein: the non-linear mapping method for the respiratory fluctuation index includes:

[0038] Calculate the difference between the respiratory fluctuation index and a preset fluctuation threshold; normalize the difference using a preset scale parameter to obtain a relative deviation; input the relative deviation into the hyperbolic tangent function to obtain a modulation amount within the range of [-1, 1]; multiply the modulation amount by a preset modulation coefficient and then add 1 to obtain a mapping coefficient; multiply the mapping coefficient by the respiratory fluctuation index to obtain the non-linearly mapped respiratory fluctuation index;

[0039] The method for calculating the dynamic weight of the body state parameter includes:

[0040] Calculate the difference between the body state parameter and a preset reference state value; multiply the difference by a preset steepness coefficient and input it into the Sigmoid function to obtain a basic weight; calculate the time change rate of the body state parameter; multiply the absolute value of the time change rate by a preset change rate influence coefficient and then add 1 to obtain a dynamic adjustment coefficient; multiply the basic weight by the dynamic adjustment coefficient to obtain the dynamic weight of the body state parameter.

[0041] As a preferred embodiment of the method for assessing and predicting the risk of childhood asthma attacks according to the present invention, wherein: the determination of the correction coefficient includes:

[0042] Determine the most relevant historical data through non-linear weight allocation according to the age characteristics of children, the calibrated risk score and its dynamic change characteristics;

[0043] Based on the selected historical data, dynamically adjust the reference value according to the change of the current risk score;

[0044] Compare the calibrated risk score with the dynamic reference to calculate the final correction coefficient;

[0045] The correction coefficient includes a static correction amount and a dynamic adjustment amount;

[0046] The calculation of the static correction amount includes:

[0047] Obtain the difference between the calibrated respiratory signal and the reference function; divide the difference by a preset reference standard deviation to obtain a normalized deviation; input the normalized deviation into the hyperbolic tangent function to obtain a static correction value within the range of [-1, 1]; multiply the static correction value by a preset main correction coefficient to obtain the final static correction amount;

[0048] The calculation of the dynamic adjustment amount includes:

[0049] Calculate the derivative of the calibrated respiratory signal with respect to time; obtain the absolute value of the derivative; divide the absolute value by the sum of the absolute value and 1 to obtain a normalized change rate within the range of [0, 1]; perform a square root operation on the normalized change rate; multiply the operation result by a preset dynamic adjustment coefficient and then add 1 to obtain the dynamic adjustment amount;

[0050] Multiply the static correction amount by the dynamic adjustment amount and then add 1 to obtain the correction coefficient.

[0051] As a preferred embodiment of the method for assessing and predicting the risk of childhood asthma attack according to the present invention, wherein: the method for dynamically adjusting the reference value includes:

[0052] Establish a static reference term, including: using a preset reference mean as the basic reference value;

[0053] Establish a periodic modulation term, including: converting the current time into a phase angle with a 24-hour period; inputting the phase angle into the sine function; multiplying the calculation result of the sine function by the product of a preset reference standard deviation and a preset periodic modulation coefficient to obtain a periodic modulation amount;

[0054] Establish a dynamic compensation term, including: obtaining the calibrated respiratory signal at the current moment; obtaining the calibrated respiratory signal at the previous moment; calculating the difference between the calibrated respiratory signals at the current moment and the previous moment; multiplying the difference by a preset change rate influence coefficient to obtain a dynamic compensation amount;

[0055] Add the basic reference value, the periodic modulation amount, and the dynamic compensation amount to obtain the dynamically adjusted reference value.

[0056] In a second aspect, the present invention provides a system for assessing and predicting the risk of childhood asthma attack, which includes:

[0057] A collection module, configured to collect the respiratory rate data, heart rate data, body temperature data, and environmental humidity data of a child, and input the respiratory rate data, the heart rate data, the body temperature data, and the environmental humidity data into a preset data processing module;

[0058] A data processing module, configured to perform fluctuation analysis on the respiration rate data based on the data processing module to obtain a respiration fluctuation index, and at the same time perform correlation analysis on the heart rate data and the body temperature data to obtain body state parameters;

[0059] An initial scoring module, configured to perform multi-dimensional cross-calculation on the respiration fluctuation index, the body state parameters and the environmental humidity data to generate an initial asthma risk score;

[0060] A correction module, configured to retrieve a reference data set matching the age of the child from a preset historical database, and perform comparative analysis on the initial asthma risk score and the reference data set to obtain a correction coefficient;

[0061] A prediction module, configured to multiply the initial asthma risk score by the correction coefficient to generate a final risk prediction value, and determine the asthma attack risk level according to the numerical range of the final risk prediction value.

[0062] The beneficial effects of the present invention are as follows: By comprehensively collecting the respiration rate, heart rate, body temperature and environmental humidity data of children, and combining data processing and analysis methods, it is possible to monitor the physical condition of children and environmental factors in real time, and accurately evaluate the asthma attack risk. Through fluctuation analysis, correlation analysis and multi-dimensional cross-calculation, a respiration fluctuation index and body state parameters are generated. Further, in combination with the influence of environmental humidity, a dynamic risk assessment model is constructed to achieve personalized risk prediction. This method can dynamically adjust the risk score, provide accurate risk level prediction, thereby providing a scientific basis for children's asthma management, effectively preventing asthma attacks, and improving the accuracy and timeliness of children's health management. BRIEF DESCRIPTION OF THE DRAWINGS

[0063] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the following drawings are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings. Among them:

[0064] Figure 1 It is a flowchart of a method for assessing and predicting the risk of asthma attacks in children. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0065] In order to make the above objects, features and advantages of the present invention more obvious and understandable, the following will describe the specific embodiments of the present invention in detail with reference to the drawings of the specification. Obviously, the described embodiments are some embodiments of the present invention, rather than all embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the scope of protection of the present invention.

[0066] In the following description, numerous specific details are set forth in order to provide a thorough understanding of the present invention. However, the present invention may be practiced in other ways than those specifically described herein. Those skilled in the art can make similar extensions without departing from the spirit of the present invention. Therefore, the present invention is not limited by the specific embodiments disclosed below.

[0067] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The appearances of "in one embodiment" in different places in this specification do not all refer to the same embodiment, nor are they separate or alternative embodiments that exclude each other.

[0068] The present invention is described in detail with reference to the schematic diagrams. When describing the embodiments of the present invention in detail, for the sake of clarity, the cross-sectional views showing the device structures are enlarged locally not in accordance with the general scale, and the schematic diagrams are only examples, which should not limit the scope of protection of the present invention. In addition, in actual production, three-dimensional spatial dimensions including length, width, and depth should be included.

[0069] At the same time, in the description of the present invention, it should be noted that the orientation or positional relationships indicated by terms such as "upper, lower, inner, and outer" are based on the orientation or positional relationships shown in the drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of the present invention. In addition, the terms "first, second, or third" are only used for descriptive purposes and cannot be construed as indicating or implying relative importance.

[0070] Unless otherwise clearly defined and limited in the present invention, the terms "mounted, connected, and coupled" should be understood in a broad sense. For example, it can be a fixed connection, a detachable connection, or an integral connection; it can also be a mechanical connection, an electrical connection, or a direct connection, or can be indirectly connected through an intermediate medium, or can be the communication inside two elements. For those of ordinary skill in the art, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0071] Example 1: Refer to Figure 1 , which is the first embodiment of the present invention. This embodiment provides a method for assessing and predicting the risk of childhood asthma attacks, including:

[0072] S1: Collect the respiratory rate data, heart rate data, body temperature data, and environmental humidity data of children, and input the respiratory rate data, heart rate data, body temperature data, and environmental humidity data into a preset data processing module;

[0073] Collect the respiratory rate data, heart rate data, body temperature data, and environmental humidity data of children. The respiratory rate data is collected every 10 minutes by a non-contact optoelectronic sensor installed at the head of the child's bed. The optoelectronic sensor is composed of an infrared emitter and a receiver with a wavelength of 940nm. The heart rate data is continuously monitored and collected by a smart bracelet with Bluetooth transmission function. The smart bracelet uses photoplethysmography to obtain the heart rate signal. The body temperature data is collected by an adhesive temperature sensor. The adhesive temperature sensor uses a medical-grade thermistor with a temperature measurement accuracy of ±0.1°C. The environmental humidity data is collected by a digital humidity sensor installed indoors. The sampling frequency of the digital humidity sensor is once every 5 minutes, and the measurement range is 0-100%RH. The respiratory rate data, heart rate data, body temperature data, and environmental humidity data are transmitted to a preset data processing module in real time through a wireless transmission module. The data processing module marks the received data with a timestamp and stores it. The data processing module includes a data acquisition unit, a data storage unit, and a data cleaning unit. The data cleaning unit screens and eliminates abnormal data.

[0074] S2: Based on the data processing module, perform fluctuation analysis on the respiratory rate data to obtain a respiratory fluctuation index. At the same time, perform correlation analysis on the heart rate data and the body temperature data to obtain body state parameters.

[0075] Based on the data processing module, perform fluctuation analysis on the respiratory rate data. The data processing module uses the sliding window method to segment the respiratory rate data. The time length of the sliding window is 30 minutes, and the window sliding step is 5 minutes.

[0076] Calculate the standard deviation of the respiratory rate data within each sliding window Calculate the mean value of the respiratory rate data within all sliding windows And calculate the respiratory rate volatility More specifically, calculate the basic fluctuation level through the relationship between the standard deviation and the mean value, and adjust the basic fluctuation using the dynamic change characteristics of the respiratory state. When the respiratory state changes rapidly, it indicates that the respiratory system is in an unstable state. At this time, the calculation result of the volatility will be increased accordingly, as shown in the following formula:

[0077] ;

[0078] Where Is the standard deviation of the respiratory rate of a single sliding window, Is the exponentially weighted moving average of the respiratory rate, Is the change rate influence coefficient, Is the change rate of the mean value;

[0079] The respiratory fluctuation rate within 24 consecutive hours is cumulatively summed to obtain the respiratory fluctuation index , more specifically, the influence factor of circadian rhythm is introduced in the calculation process, so that the volatility in different time periods has different weights, as shown in the following formula:

[0080] ;

[0081] Among them, is the circadian rhythm regulation coefficient 0.15, is the time series index, is the th respiratory fluctuation rate at the time point, is the physiological rhythm fluctuation simulating a 24-hour cycle.

[0082] At the same time, the heart rate data and the body temperature data are subjected to correlation analysis, and the correlation analysis includes: standardizing the heart rate data according to age characteristics to obtain the heart rate standard score , the calculation of the heart rate standard score adopts the calculation method of (measured heart rate - basal heart rate) / age coefficient, calculates based on the deviation degree between the measured heart rate and the basal heart rate, and considers the influence of age factors. At the same time, the dynamic characteristics of heart rate change are introduced as a regulatory factor. When the heart rate changes rapidly, it indicates that cardiovascular regulation is abnormal, and at this time, the standard score will be adjusted accordingly. This calculation method not only considers the static deviation of the heart rate level but also includes the dynamic characteristics of heart rate change, and can more comprehensively reflect the cardiovascular function state, as shown in the following formula:

[0083] ;

[0084] Among them, is the measured heart rate, is the basal heart rate, is the age coefficient, is the dynamic regulation coefficient 0.25, is the heart rate change rate.

[0085] The body temperature data is converted into a body temperature change curve, the peak value and the trough value of the body temperature change curve are extracted, and the temperature difference value is calculated, as shown in the following formula:

[0086] ;

[0087] Among them, is the temperature acceleration influence coefficient 0.3, is the second derivative of temperature, is the peak temperature of body temperature fluctuation, is the valley temperature of body temperature fluctuation.

[0088] Substitute the heart rate standard score and the temperature difference value into a preset state evaluation equation for calculation, where and are weight coefficients obtained from clinical data, and the calculation result of the state evaluation equation is the body state parameter , as shown in the following formula:

[0089] ;

[0090] Among them, is the heart rate weight coefficient 0.4, is the temperature difference weight coefficient 0.3, is the interaction term coefficient 0.2, is the comprehensive influence coefficient 0.1.

[0091] S3: Perform multi-dimensional cross-calculation on the respiratory fluctuation index, the body state parameter, and the environmental humidity data to generate an initial asthma risk score;

[0092] Extract the time series characteristics of the environmental humidity data, and construct a humidity influence model using an improved Gaussian kernel function group. The time response characteristics of the Gaussian kernel function match the physiological response characteristics of children's respiratory tract to humidity changes.

[0093] This function is used to describe the dynamic change characteristics of environmental humidity over time. First, a reference humidity value is set as the basic level, and then multiple Gaussian functions are superimposed to simulate the humidity fluctuations at different time points. Each Gaussian function has its characteristic time point, fluctuation amplitude, and time scale parameter, and these parameters together determine the position, intensity, and duration of the fluctuation. In this way, complex humidity change patterns, including sudden changes and persistent fluctuations, can be accurately depicted. As shown in the following formula:

[0094] ;

[0095] Among them, is the reference humidity value, is the humidity fluctuation amplitude, is the current time point, is the total number of characteristic time points, is the characteristic time point, is the humidity influence model, is the time scale parameter, which is dynamically adjusted by an adaptive algorithm:

[0096] ;

[0097] Among them, is the reference time scale of 30 minutes, is the adaptation coefficient of 0.3, represents the humidity value calculated at the characteristic time point .

[0098] Calculate the dynamic sensitivity index based on the humidity impact model , and the dynamic sensitivity index reflects the degree of stimulation of environmental humidity changes on children's respiratory tracts. This index evaluates the dynamic characteristics of humidity changes. First, calculate the rate of change of humidity over time (first derivative) to reflect the speed of humidity change. At the same time, introduce the acceleration of humidity change (second derivative) as a regulatory factor. When the humidity change accelerates or decelerates, adjust the sensitivity through the acceleration impact factor. Integrate this comprehensive effect over the entire observation period and normalize it through the observation time period to finally obtain the sensitivity index reflecting the dynamic change characteristics of humidity. As shown in the following formula:

[0099] ;

[0100] Among them, is the observation time period, set to 6 hours, is the acceleration impact factor of 0.25; the dynamic sensitivity index is updated every 30 minutes and a moving average is performed with historical data:

[0101] ;

[0102] Among them, is the smoothed dynamic sensitivity index at the current time point , is the smoothed dynamic sensitivity index at the previous time point , is the original dynamic sensitivity index at the current time point , is the smoothing coefficient of 0.3.

[0103] Perform a non-linear mapping on the respiratory fluctuation index RVI to enhance the response sensitivity to abnormal fluctuations, specifically including:

[0104] Calculate the difference between the respiratory fluctuation index and the preset fluctuation threshold; normalize the difference using the preset scale parameter to obtain the relative deviation; input the relative deviation into the hyperbolic tangent function to obtain the modulation amount within the range of [-1, 1]; multiply the modulation amount by the preset modulation coefficient and then add 1 to obtain the mapping coefficient; multiply the mapping coefficient by the respiratory fluctuation index to obtain the non-linearly mapped respiratory fluctuation index, as shown in the following formula:

[0105] ;

[0106] Among them, is the modulation coefficient of 0.3, is the fluctuation threshold, is the scale parameter, is the respiratory fluctuation index mapping parameter, which is set by age segment:

[0107] Under 3 years old: = 15;

[0108] 3 - 6 years old: = 12;

[0109] 6 - 12 years old: = 10.

[0110] Construct the dynamic weight function of the body state parameter to achieve adaptive weighting for different state intervals. More specifically, the calculation method of the dynamic weight includes:

[0111] Calculate the difference between the body state parameter and the preset reference state value; multiply the difference by the preset steepness coefficient and input it into the Sigmoid function to obtain the basic weight; calculate the time change rate of the body state parameter; multiply the absolute value of the time change rate by the preset change rate influence coefficient and add 1 to obtain the dynamic adjustment coefficient; multiply the basic weight by the dynamic adjustment coefficient to obtain the dynamic weight of the body state parameter, as shown in the following formula:

[0112] ;

[0113] Among them, is the steepness coefficient, with a basic value of 2.0 and increasing by 0.1 with age; is the reference state value, obtained by statistical analysis of individual historical data; is the change rate influence coefficient of 0.2; is the dynamic weight function of the body state parameter, and the dynamic weight function is updated every 15 minutes;

[0114] Substitute the above parameters into the multi - dimensional cross - evaluation equation to calculate the initial asthma risk score RS through polynomial combination:

[0115] ;

[0116] Among them, is the learning rate parameter in the optimization algorithm, and the weight coefficients to are dynamically adjusted using an adaptive optimization algorithm:

[0117] ;

[0118] Among them, is the historical prediction error of each index, is the learning rate parameter in the optimization algorithm, is the adaptive dynamic weight.

[0119] Furthermore, considering the daily cycle change characteristics, the initial asthma risk score RS is dynamically calibrated:

[0120] ;

[0121] Among them, is the calibrated risk score, is the daily cycle adjustment coefficient 0.1, is the acceleration influence coefficient 0.15; The calibrated risk score is divided according to the following intervals:

[0122] 0 - 0.3: Low risk

[0123] 0.3 - 0.6: Medium risk

[0124] 0.6 - 0.8: High risk

[0125] 0.8 - 1.0: Extremely high risk.

[0126] S4: Retrieve a reference data set matching the age of the child from a preset historical database, and compare and analyze the initial asthma risk score with the reference data set to obtain a correction coefficient;

[0127] First, according to the age characteristics of the child, the calibrated risk score and its dynamic change characteristics, the most relevant historical data are determined through non - linear weight allocation, as shown in the following formula:

[0128] ;

[0129] Among them, represents the matching degree between historical data and current data, is the age value of the sample in the database, is the actual age of the target child, is the standard deviation parameter for age matching, controlling the tolerance range of age matching, is the calibrated risk score at the current moment, is the benchmark risk score in the historical data, is the coupling coefficient (0.3), controlling the influence intensity of the risk score on the matching degree, is the dynamic response coefficient (0.2), adjusting the sensitivity of the system to the change rate of the risk score, is the time change rate of the risk score, is the natural exponential function.

[0130] Furthermore, based on the selected historical data, combined with the daily cycle characteristics and the change trend of the risk score, through an adaptive algorithm, the reference value is dynamically adjusted according to the change of the current risk score, so that the evaluation result is more in line with the individual characteristics. Among them, the adaptive algorithm is shown as follows:

[0131] ;

[0132] Among them, is the average risk score of the reference group, is the standard deviation of the reference group, is the cycle modulation coefficient (0.15), which controls the amplitude of the daily cycle fluctuation, is the change rate influence coefficient (0.25), which determines the influence degree of the risk score change on the reference, is the time interval, which is used to calculate the score change, is the current time, is the adaptive reference benchmark, is the calibrated risk score at the current moment, is the calibrated risk score at the moment.

[0133] Furthermore, the calibrated risk score is compared with the dynamic reference, and the final correction coefficient is calculated by considering the score deviation degree and the change trend. The correction coefficient includes a static correction amount and a dynamic adjustment amount;

[0134] The calculation of the static correction amount includes:

[0135] Obtain the difference between the calibrated respiratory signal and the reference function; divide the difference by the preset reference standard deviation to obtain the normalized deviation; input the normalized deviation into the hyperbolic tangent function to obtain the static correction value in the interval [-1, 1]; multiply the static correction value by the preset main correction coefficient to obtain the final static correction amount;

[0136] The calculation of the dynamic adjustment amount includes:

[0137] Calculate the derivative of the calibrated respiratory signal with respect to time; obtain the absolute value of the derivative; divide the absolute value by the sum of the absolute value and 1 to obtain the normalized change rate in the interval [0, 1]; perform a square root operation on the normalized change rate; multiply the operation result by the preset dynamic adjustment coefficient and then add 1 to obtain the dynamic adjustment amount;

[0138] Multiply the static correction amount by the dynamic adjustment amount and then add 1 to obtain the correction coefficient, as shown in the following formula:

[0139] ;

[0140] Among them, is the main correction coefficient (0.35), which controls the intensity of the overall correction, is the dynamic adjustment coefficient (0.2), which adjusts the influence of the change rate on the correction, is the hyperbolic tangent function, which is used to limit the correction value within a reasonable range, is the standardized scoring deviation, is the normalized change rate term, is the correction coefficient.

[0141] S5: Multiply the initial asthma risk score by the correction coefficient to generate a final risk prediction value, and determine the asthma attack risk level according to the numerical range of the final risk prediction value.

[0142] Perform an exact multiplication operation on the initial asthma risk score and the correction coefficient. Based on the 32-bit floating-point number operation standard, the output accuracy is maintained to 4 decimal places; the result of the multiplication operation is used as the initial value of the risk prediction, and the calculation result is updated every 5 minutes.

[0143] According to medical expert experience and clinical statistical data, the smoothed risk prediction value is divided into five risk level intervals:

[0144] Safety level: The final risk prediction value ≤ 0.2 indicates a very low risk of asthma attack, and normal daily routines can be maintained;

[0145] Warning level: 0.2 < the final risk prediction value ≤ 0.4 indicates a potential risk, and the monitoring frequency should be appropriately increased;

[0146] Early warning level: 0.4 < the final risk prediction value ≤ 0.6 indicates a significantly increased risk, and preventive medication should be taken;

[0147] High-risk level: The final risk prediction value > 0.6 indicates a high risk of attack. It is recommended to take preventive medications immediately and make first aid preparations.

[0148] According to the characteristics of children of different ages, the thresholds of the risk levels are dynamically adjusted:

[0149] For children under 3 years old, considering the imperfect development of their respiratory tracts, the thresholds of each level are lowered by 0.05 as a whole;

[0150] For children aged 3 - 6 years old, the standard threshold settings are maintained;

[0151] For children aged 6 - 12 years old, according to their immune capacity, the thresholds can be appropriately increased by 0.05.

[0152] Furthermore, this embodiment also provides a system for assessing and predicting the risk of childhood asthma attacks, including:

[0153] An acquisition module, configured to acquire the respiratory rate data, heart rate data, body temperature data, and environmental humidity data of a child, and input the respiratory rate data, the heart rate data, the body temperature data, and the environmental humidity data into a preset data processing module;

[0154] A data processing module, configured to perform fluctuation analysis on the respiratory rate data based on the data processing module to obtain a respiratory fluctuation index, and at the same time perform correlation analysis on the heart rate data and the body temperature data to obtain a physical state parameter;

[0155] An initial scoring module, configured to perform multi-dimensional cross-calculation on the respiratory fluctuation index, the physical state parameter, and the environmental humidity data to generate an initial asthma risk score;

[0156] A correction module, configured to retrieve a reference data set matching the age of the child from a preset historical database, and perform comparative analysis on the initial asthma risk score and the reference data set to obtain a correction coefficient;

[0157] A prediction module, configured to multiply the initial asthma risk score by the correction coefficient to generate a final risk prediction value, and determine the asthma attack risk level according to the numerical range of the final risk prediction value.

[0158] This embodiment also provides a computer device, applicable to the situation of the method for assessing and predicting the risk of childhood asthma attacks, including a memory and a processor; the memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions to implement the method for assessing and predicting the risk of childhood asthma attacks as proposed in the above embodiment.

[0159] The computer device can be a terminal, which includes a processor, a memory, a communication interface, a display screen, and an input device connected by a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage medium. The communication interface of the computer device is used to communicate with external terminals in a wired or wireless manner, and the wireless manner can be implemented through WIFI, carrier networks, NFC (Near Field Communication), or other technologies. The display screen of the computer device can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device can be a touch layer covering the display screen, or buttons, trackballs, or touchpads provided on the housing of the computer device, or an external keyboard, touchpad, or mouse, etc.

[0160] This embodiment also provides a storage medium, on which a computer program is stored, and when the program is executed by a processor, it implements the method for evaluating and predicting the risk of childhood asthma attacks proposed in the above embodiment.

[0161] The storage medium proposed in this embodiment and the data storage method proposed in the above embodiment belong to the same inventive concept. For technical details not described in detail in this embodiment, reference can be made to the above embodiment, and this embodiment has the same beneficial effects as the above embodiment.

[0162] Embodiment 2: This is the second embodiment of the present invention. This embodiment provides a method for evaluating and predicting the risk of childhood asthma attacks. In order to verify the beneficial effects of the present invention, scientific demonstration is carried out through simulation experiments.

[0163] To verify the effectiveness of the childhood asthma risk warning method proposed by the present invention, 60 asthma children in the outpatient department of the respiratory department of a certain children's specialized hospital were selected for a 3-month follow-up monitoring study. The ages of the research objects were distributed between 2 and 10 years old, including 15 children under 3 years old, 25 children between 3 and 6 years old, and 20 children between 6 and 12 years old. All research objects were diagnosed as mild to moderate asthma by a specialist doctor and had at least one acute attack within 3 months before the start of the experiment.

[0164] The preparatory work before the experiment included: equipping each child with a set of standardized monitoring devices, including a non-contact photoelectric sensor (sampling accuracy ±2 times / minute) using an infrared emitter and receiver tube with a wavelength of 940nm, a smart bracelet with Bluetooth transmission function (heart rate sampling frequency 60Hz), a medical-grade thermistor attached temperature sensor (model: temperature measurement accuracy ±0.1°C), and a digital humidity sensor (measurement range 0-100%RH, accuracy ±2%RH). All devices were professionally calibrated to ensure the accuracy of data collection.

[0165] The test experiment was carried out according to the steps of Example 1 to obtain the following experimental data:

[0166] Table 1. Data analysis of clinical trial of childhood asthma risk early warning system

[0167] Age group Sample size Early warning accuracy rate (%) Omission rate (%) False alarm rate (%) Average early warning lead time (hours) Accuracy rate of traditional method (%) System response time (minutes) 2 - 3 year old group 15 92.5 3.2 4.3 4.8 75.3 2.5 3 - 4 year old group 12 93.1 2.8 4.1 5.2 76.8 2.3 4 - 5 year old group 8 94.2 2.5 3.3 5.5 77.5 2.2 5 - 6 year old group 5 94.8 2.3 2.9 5.8 78.2 2.1 6 - 8 year old group 12 95.3 2.1 2.6 6.2 79.1 2.0 8 - 10 year old group 8 96.1 1.8 2.1 6.5 80.2 1.8

[0168] The analysis results based on the above experimental data show that the childhood asthma risk warning method proposed in the present invention has significant clinical application value. First of all, in terms of warning accuracy, the system showed excellent performance in all age groups, and the overall accuracy was maintained above 92.5%. Especially in the 8-10 year old group, the accuracy reached 96.1%, which is mainly due to the multi-dimensional cross-calculation method adopted by the present invention, which effectively integrates multiple key physiological indicators such as respiratory rate, heart rate, body temperature and environmental humidity.

[0169] From the perspective of missed alarm rate and false alarm rate, the system shows extremely high reliability. Even in the 2-3 year old age group with large fluctuations in physiological indicators, the missed alarm rate and false alarm rate are only 3.2% and 4.3% respectively, which are much lower than traditional monitoring methods. This advantage mainly comes from the improved Gaussian kernel function group and adaptive optimization algorithm adopted by the present invention, which can accurately capture the changing laws of physiological characteristics of children of different age groups.

[0170] The warning lead time is an important indicator for evaluating the effectiveness of the warning system. Data show that the system can issue warnings 4.8-6.5 hours in advance, which provides a sufficient time window for clinical intervention. It is worth noting that with the increase of age, the warning lead time tends to be gradually extended, which is related to the more typical symptoms and more stable physiological indicators of older children.

[0171] Compared with traditional early warning methods, the present invention has shown significant advantages in all age groups. The accuracy of traditional methods is between 75.3% and 80.2%, while the early warning accuracy of the present invention is generally 15-20 percentage points higher. This advantage is mainly reflected in: 1) Real-time collection and analysis of multi-dimensional data, avoiding the limitations brought by a single indicator; 2) Innovative introduction of environmental humidity factors to establish a more complete risk assessment model; 3) Adaptive algorithms are used to personalize modeling of the physiological characteristics of children of different age groups.

[0172] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A method for assessing and predicting the risk of childhood asthma attacks, characterized in that: Including: Collecting the respiratory rate data, heart rate data, body temperature data and environmental humidity data of children, and inputting the respiratory rate data, the heart rate data, the body temperature data and the environmental humidity data into a preset data processing module; Based on the data processing module, performing fluctuation analysis on the respiratory rate data to obtain a respiratory fluctuation index, and at the same time performing correlation analysis on the heart rate data and the body temperature data to obtain body state parameters; Performing multi-dimensional cross-calculation on the respiratory fluctuation index, the body state parameters and the environmental humidity data to generate an initial asthma risk score; Retrieving a reference data group matching the age of the child from a preset historical database, and performing comparative analysis on the initial asthma risk score and the reference data group to obtain a correction coefficient; Multiplying the initial asthma risk score by the correction coefficient to generate a final risk prediction value, and determining the asthma attack risk level according to the numerical range of the final risk prediction value; The calculation of the respiratory fluctuation index is shown in the following formula: ； Wherein, ； wherein, is the standard deviation of the respiratory rate for a single time window, is the exponentially moving average of the respiratory rate, is the change rate influence coefficient, is the change rate of the mean value, is the circadian rhythm regulation coefficient, is the time series index, is the respiratory rate volatility at the is the physiological rhythm fluctuation simulating a 24-hour cycle; The correlation analysis to obtain the body state parameters is shown in the following formula: ； Among them, is the heart rate weight coefficient 0.4, is the temperature difference weight coefficient 0.3, is the interaction term coefficient 0.2, is the comprehensive influence coefficient; wherein, the standard heart rate score value is obtained as shown in the following formula: ； Among them, is the measured heart rate, is the basal heart rate, is the age coefficient, is the dynamic adjustment coefficient, is the heart rate change rate; wherein, the temperature difference value is obtained as shown in the following formula: ； Among them, is the temperature acceleration influence coefficient 0.3, is the second derivative of temperature, is the peak temperature of body temperature fluctuation, is the trough temperature of body temperature fluctuation; The initial asthma risk score is calculated by polynomial combination, as shown in the following formula: ； Among them, it is the learning rate parameter in the optimization algorithm, and the weight coefficient to is dynamically adjusted by using an adaptive optimization algorithm: ； wherein, is the historical prediction error of each index, is the learning rate parameter in the optimization algorithm, is the adaptive dynamic weight; Among them, is the dynamic weight function of the body state parameter, as shown in the following formula: ； Among them, is the steepness coefficient, which increases with age; is the reference state value, is the change rate influence coefficient; Among them, The dynamic sensitivity index, which is calculated based on the humidity influence model, is shown in the following formula: ； Among them, is the observation time period, is the humidity influence model, is the acceleration influence factor 0.25; Wherein, the humidity influence model is shown in the following formula: ； Among them, is the reference humidity value, is the humidity fluctuation amplitude, is the current time point, is the total number of characteristic time points, is the characteristic time point, is the time scale parameter; The dynamic sensitivity index is updated every 30 minutes and is subjected to a moving average with historical data: ； Among them, is the smoothed dynamic sensitivity index at the current time point , is the smoothed dynamic sensitivity index at the previous time point , is the original dynamic sensitivity index at the current time point , is the smoothing coefficient; Among them, the is the respiratory fluctuation index mapping parameter, as shown in the following formula: ； Among them, is the modulation coefficient, is the fluctuation threshold, is the scale parameter.

2. The method for assessing and predicting the risk of childhood asthma attacks according to claim 1, wherein: The determination of the correction coefficient includes: According to the age characteristics of children, the calibrated risk score and its dynamic change characteristics, determining the most relevant historical data through non-linear weight allocation; Based on the selected historical data, dynamically adjusting the reference value according to the change of the current risk score; Comparing the calibrated risk score with the dynamic reference to calculate the final correction coefficient; The correction coefficient includes a static correction amount and a dynamic adjustment amount; The calculation of the static correction amount includes: Obtaining the difference between the calibrated respiratory signal and the reference function; dividing the difference by a preset reference standard deviation to obtain a normalized deviation; inputting the normalized deviation into a hyperbolic tangent function to obtain a static correction value within the range of [-1, 1]; multiplying the static correction value by a preset main correction coefficient to obtain the final static correction amount; The calculation of the dynamic adjustment amount includes: Calculating the derivative of the calibrated respiratory signal with respect to time; obtaining the absolute value of the derivative; dividing the absolute value by the sum of the absolute value and 1 to obtain a normalized change rate within the range of [0, 1]; performing a square root operation on the normalized change rate; multiplying the operation result by a preset dynamic adjustment coefficient and then adding 1 to obtain the dynamic adjustment amount; Multiplying the static correction amount by the dynamic adjustment amount and then adding 1 to obtain the correction coefficient.

3. The method for assessing and predicting the risk of childhood asthma attacks according to claim 2, characterized in that: The method for dynamically adjusting the reference value includes: Establishing a static reference term, including: using a preset reference mean as the basic reference value; Establishing a periodic modulation term, including: converting the current time into a phase angle with a 24-hour period; inputting the phase angle into a sine function; multiplying the calculation result of the sine function by the product of a preset reference standard deviation and a preset periodic modulation coefficient to obtain a periodic modulation amount; Establish a dynamic compensation term, including: obtaining the calibrated respiratory signal at the current moment; obtaining the calibrated respiratory signal at the previous moment; calculating the difference between the calibrated respiratory signals at the current moment and the previous moment; multiplying the difference by a preset change rate influence coefficient to obtain a dynamic compensation amount; Adding the basic reference value, the periodic modulation amount, and the dynamic compensation amount to obtain a dynamically adjusted reference value.

4. A child asthma attack risk assessment and prediction system, based on the child asthma attack risk assessment and prediction method according to any one of claims 1 to 3, characterized in that: Including: A collection module for collecting the respiratory frequency data, heart rate data, body temperature data, and environmental humidity data of a child, and inputting the respiratory frequency data, heart rate data, body temperature data, and environmental humidity data into a preset data processing module; A data processing module for performing fluctuation analysis on the respiratory frequency data based on the data processing module to obtain a respiratory fluctuation index, and at the same time performing correlation analysis on the heart rate data and the body temperature data to obtain body state parameters; An initial scoring module for performing multi-dimensional cross-calculation on the respiratory fluctuation index, the body state parameters, and the environmental humidity data to generate an initial asthma risk score; A correction module for retrieving a reference data group matching the age of the child from a preset historical database, and performing comparative analysis on the initial asthma risk score and the reference data group to obtain a correction coefficient; A prediction module for multiplying the initial asthma risk score by the correction coefficient to generate a final risk prediction value, and determining the asthma attack risk level according to the numerical range of the final risk prediction value.

Images