Child health state comprehensive evaluation management method and system based on family-garden cooperative rearing

Abstract

The embodiment of the invention provides an infant health state comprehensive evaluation management method and system based on family cooperative rearing, and the method comprises the steps: obtaining infant health state data and family environment health factor data, and the environment health factors comprise air humidity and temperature; based on the infant health state data and the environmental health factor data, the infant respiratory disease risk is evaluated; based on the respiratory disease risk assessment result, generating an environment regulation and control strategy to optimize home environment setting; and the environment regulation and control strategy is implemented, and the air humidity and temperature in the home environment are adjusted. Through the scheme of the embodiment of the invention, the problem of how to regulate and control home environment setting according to environmental health factors (such as air humidity and temperature) so as to solve the problem of susceptibility of children to respiratory diseases can be solved.

Description

Technical Field

[0001] This application relates to health information technology, and specifically to a comprehensive assessment and management method and system for the health status of young children based on home-school co-education. Background Art

[0002] The comprehensive assessment and management method and system for the health status of young children based on home-school co-education is a comprehensive platform that integrates the collaboration between families and kindergartens. It aims to evaluate and manage the overall health status of young children through data collection and analysis (such as children's physiological indicators, behavioral habits, etc.), thereby optimizing parenting strategies. However, this system has a key problem: how to dynamically adjust home environment settings (such as indoor ventilation or humidification equipment) according to environmental health factors (such as air humidity and temperature) to effectively address the common risks of young children being susceptible to respiratory diseases. This involves the challenge of integrating real-time monitoring and intelligent regulation. Summary of the Invention

[0003] In view of this, the embodiments of the present disclosure provide a comprehensive assessment and management method and system for the health status of young children based on home-school co-education, which at least partially solves the problems existing in the prior art.

[0004] A comprehensive assessment and management method and system for children's health status based on home-school co-education, including:

[0005] Acquiring child health status data and home environment health factor data, wherein the environmental health factors include air humidity and temperature;

[0006] Assessing the risk of respiratory diseases in young children based on the health status data and environmental health factor data of the young children;

[0007] Based on the respiratory disease risk assessment results, generate environmental control strategies to optimize home environment settings;

[0008] Implement the environmental control strategy to adjust the air humidity and temperature in the home environment.

[0009] According to one embodiment, the step of assessing the risk of respiratory diseases in young children further comprises:

[0010] Get the child's age data Age, where Age represents the child's age (unit: years);

[0011] Get real-time air humidity H and air temperature T, where H represents air humidity (unit: percentage) and T represents air temperature (unit: degrees Celsius);

[0012] The risk score is calculated based on the following formula: Risk_score = a × (1 / H) + b × |T25| + c × (1 / Age), where a, b, and c are preset weight coefficients (a represents the humidity impact weight, b represents the temperature deviation weight, and c represents the age impact weight). This formula quantifies risk: the lower the humidity, the greater the temperature deviation from 25°C, or the younger the age, the higher the risk.

[0013] If Risk_score>risk threshold R_th (R_th is a preset threshold), it is judged as high risk; otherwise it is judged as low risk.

[0014] According to one embodiment, the step of assessing the risk of respiratory diseases in young children further comprises:

[0015] Get the child's weight data Weight, where Weight represents the child's weight (unit: kilogram);

[0016] Adjust the risk score based on weight data: Risk_score = Risk_score × (1 + k_w × (1 / Weight)), where k_w is the weight adjustment factor (k_w>0). This adjustment means that the lighter the weight, the greater the risk score increase;

[0017] If the adjusted Risk_score>R_th, it is judged as high risk; otherwise, it is judged as low risk;

[0018] Generate a risk level report based on the determination results.

[0019] According to one embodiment, the step of assessing the risk of respiratory diseases in young children further comprises:

[0020] Obtain historical disease frequency data Freq, where Freq represents the number of respiratory diseases that occurred in young children in the past year;

[0021] Update the risk score based on the following formula: Risk_score = Risk_score + d × Freq, where d is the historical disease weight coefficient (d>0). This update means that the higher the historical disease frequency, the higher the risk score;

[0022] If the updated Risk_score>R_th, it is judged as high risk; otherwise, it is judged as low risk;

[0023] Correlate and analyze risk level reports with historical data.

[0024] According to one embodiment, generating an environmental control strategy to optimize home environment settings further includes:

[0025] Determine the target humidity H_target and the target temperature T_target based on the risk score Risk_score;

[0026] Calculate H_target = 50 + m × Risk_score and T_target = 25 + n × (Risk_score - R_th) based on the following formula, where m and n are adjustment coefficients (m < 0, n < 0). The meaning of this formula is: the higher the risk, the higher the target humidity needs to be increased (because low humidity exacerbates the risk), and the target temperature needs to approach 25°C (to reduce the deviation);

[0027] If the current humidity H < H_target, the strategy is to increase the humidity; otherwise, maintain or reduce the humidity;

[0028] If the current temperature T < T_target or T > T_target, the strategy is to adjust the temperature to T_target.

[0029] According to one embodiment, the generating of the environmental control strategy to optimize the home environment settings further includes:

[0030] Obtain the season data Season, where Season represents the current season (encoded as a numerical value, such as winter = 1, summer = 2);

[0031] Adjust the target values based on the season: H_target = H_target × S_h and T_target = T_target × S_t, where S_h and S_t are season factors (in winter, S_h = 2, S_t = 1; in summer, S_h = 8, S_t = 9). This adjustment means that higher humidity and temperature are required in winter to reduce the impact of dryness and cold;

[0032] If Season is winter and H < H_target, the strategy preferentially increases the humidity;

[0033] Output the optimized environmental control strategy.

[0034] According to one embodiment, the implementing of the environmental control strategy further includes:

[0035] Real-time monitor the adjusted air humidity H and air temperature T;

[0036] Calculate the control effect index E = |H - H_target| / H_target + |T - T_target| / T_target based on the following formula, where E represents the degree of deviation between the actual and the target environment (the smaller the value, the better the effect);

[0037] If E > E_th (E_th is a preset effect threshold), trigger a re-evaluation of the risk;

[0038] Dynamically update the control strategy based on the reassessment results.

[0039] According to one embodiment, the implementing environment control strategy further includes:

[0040] Introduce a feedback control loop and adjust the coefficients based on the control effect index E: if E>E_th, then m=m×9 and n=n×9 (reduce the step size to prevent over-control);

[0041] Recalculate H_target and T_target based on the updated m and n;

[0042] If E≤E_th for three consecutive times, maintain the current strategy; otherwise, continue to adjust;

[0043] Store feedback data for historical analysis.

[0044] According to one embodiment, the step of assessing the risk of respiratory diseases in young children further comprises:

[0045] Predicting a risk score using a machine learning model, wherein the model inputs include H, T, Age, and Weight;

[0046] Optimize model weights based on the following formula: Weight update Δw = η × (Risk_score_actualRisk_score_predicted) × input features, where η is the learning rate, Risk_score_actual is the actual risk score, and Risk_score_predicted is the model's predicted value. This formula means: reduce prediction error through gradient descent;

[0047] If the prediction error |Risk_score_actualRisk_score_predicted|>error threshold, retrain the model;

[0048] The optimized risk score is output and used to generate strategies.

[0049] According to one embodiment, the method of using a machine learning model to generate an environmental control strategy to optimize home environment settings further includes:

[0050] Specify the model as a neural network, the input layer includes H, T, Age, Weight, Freq, and the output layer is Risk_score;

[0051] The hidden layer is activated based on the following conditions: if the input feature normalization value is > 5, the ReLU activation function is used; otherwise, the Sigmoid activation function is used.

[0052] Train the model until the loss function \(L=(1 / N)\times\sum(Risk\_score\_actual - Risk\_score\_predicted)^2 < L\_th\) (\(L\_th\) is the loss threshold), where \(N\) is the number of samples;

[0053] Deploy the trained model for real-time risk assessment;

[0054] Add a user confirmation step: Push the regulation strategy to parents or teachers;

[0055] Adjust the strategy based on user feedback: If the feedback is consent, implement the strategy; if the feedback is modification, update \(H\_target = H\_target\times F\_h\) and \(T\_target = T\_target\times F\_t\) based on the following formula, where \(F\_h\) and \(F\_t\) are feedback factors (\(F\_h = 95\) represents a slight decrease in humidity, \(F\_t = 05\) represents a slight increase in temperature);

[0056] If no feedback is received, delay implementing the strategy until timeout;

[0057] Record user interaction data for subsequent strategy optimization.

[0058] The embodiments of the present disclosure provide a comprehensive assessment and management method and system for the health status of young children based on home-school cooperation, including: obtaining young children's health status data and home environment health factor data, where the environmental health factors include air humidity and temperature; evaluating the risk of young children's respiratory diseases based on the young children's health status data and environmental health factor data; generating an environmental regulation strategy based on the respiratory disease risk assessment result to optimize the home environment settings; implementing the environmental regulation strategy to adjust the air humidity and temperature in the home environment. Through the solution of the embodiments of the present disclosure, it is possible to solve the problem of how to regulate the home environment settings according to environmental health factors (such as air humidity, temperature) to address the problem of young children's susceptibility to respiratory diseases. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] To more clearly illustrate the technical solutions of the exemplary embodiments of the present disclosure, the following will briefly introduce the drawings required in the embodiments. It should be understood that the following drawings only show some embodiments of the present disclosure, and thus should not be regarded as limiting the scope. For those of ordinary skill in the art, other related drawings can be obtained based on these drawings without creative efforts.

[0060] Figure 1 is a flowchart of a comprehensive assessment and management method and system for the health status of young children based on home-school cooperation;

[0061] Figure 2 is a further flowchart for evaluating the risk of young children's respiratory diseases;

[0062] Figure 3 This is a further flowchart for assessing the risk of respiratory disease in young children;

[0063] Figure 4 This is a further flowchart for assessing the risk of respiratory disease in young children;

[0064] Figure 5 It is a further flow chart for generating environmental control strategies to optimize the home environment settings;

[0065] Figure 6 It is a further flow chart for generating environmental control strategies to optimize the home environment settings;

[0066] Figure 7 It is a further flow chart for implementing environmental control strategies;

[0067] Figure 8 It is a further flow chart for implementing environmental control strategies;

[0068] Figure 9 This is a further flowchart for assessing the risk of respiratory disease in young children;

[0069] Figure 10 This is a further flowchart of using machine learning models to generate environmental control strategies to optimize home environment settings. DETAILED DESCRIPTION

[0070] In order to make the objectives, technical solutions and advantages of the embodiments of the present disclosure more clear, the embodiments of the present disclosure are further described in detail below in combination with the embodiments and drawings. The schematic implementation methods of the embodiments of the present disclosure and their descriptions are only used to explain the embodiments of the present disclosure and are not intended to limit the embodiments of the present disclosure.

[0071] Next, with reference to the accompanying drawings, the present invention describes a comprehensive assessment and management method and system for early childhood health based on home-school co-education, which integrates data from home and kindergarten environments to prevent respiratory diseases in young children. The system first performs step 1: acquiring data on the child's health status and health factors in the home environment, where these factors include air humidity and temperature. Specifically, the system collects parent-reported child health data, such as body temperature, cough frequency, and allergy history, as well as activity records and health questionnaires entered by kindergarten teachers, through a mobile application or cloud platform. Simultaneously, IoT sensors (such as thermometers and hygrometers) monitor humidity and temperature data in the home and kindergarten environments in real time, and all this information is synchronized to a central database for preprocessing and storage. For example, in one embodiment, a parent uses a dedicated app to input daily symptom data for their child, Xiao Ming (3 years old with a history of asthma), such as mild nasal congestion and a low-grade fever. Simultaneously, a smart sensor in the living room detects a humidity of 35% (lower than the ideal range of 40-60%) and a temperature of 18°C. Sensors in the kindergarten classroom also upload data showing a humidity of 30% and a temperature of 20°C. The system automatically integrates this information to ensure data integrity and real-time availability, laying the foundation for subsequent analysis.

[0072] Step 2: Based on the health status data of the children and the environmental health factor data, the risk of respiratory diseases in young children is assessed. In specific operations, the system uses a machine learning algorithm (such as a decision tree or a neural network) to analyze the input data: it combines historical health records (such as the frequency of previous infections), current symptoms, and environmental factors (too low humidity or temperature fluctuations) to calculate a risk score. This score quantifies the probability of respiratory diseases (such as colds or bronchitis) based on medical guidelines (such as WHO standards); the system also takes into account seasonal factors and individual differences (such as age or physique) and outputs a visual report to facilitate parents and teachers to understand the risk level. For example, specifically for Xiao Ming's case, the system analyzes its input data: 35% humidity increases the risk of dry mucous membranes. Combined with recent cough symptoms and low temperature environment, the algorithm predicts that the risk of respiratory infection is high (score 80 / 100), and generates an early warning prompt, emphasizing that insufficient humidity is the main cause, helping both home and school to intervene in time.

[0073] Step three: Based on the respiratory disease risk assessment results, generate an environmental control strategy to optimize the home environment settings. In terms of specific operations, the system applies an optimization model (such as a PID control algorithm) to generate personalized strategies based on the risk level and specific environmental data: for example, if the risk is high and the humidity is low, the strategy recommends increasing the humidity to 50-60% and stabilizing the temperature at 20-22°C; the strategy also includes device operation instructions (such as humidifier power settings) and user suggestions (such as ventilation time), and is pushed through APP or email to ensure collaborative execution between home and school. For example, in one embodiment, for Xiao Ming's high-risk assessment, the system generates a strategy: the home environment needs to use a smart humidifier to increase the humidity to 55% and adjust the temperature to 21°C; the kindergarten environment recommends that teachers turn on the air conditioner constant temperature mode, and add strategy details such as humidification for 2 hours a day, with scientific basis (such as increased humidity can reduce the survival rate of the virus) to solve the susceptibility problem.

[0074] Step 4: Implement the environmental control strategy to adjust the air humidity and temperature in the home environment. In specific operations, the system connects to smart home devices (such as humidifiers or air conditioners) through the API interface, and automatically sends control instructions to execute the strategy; at the same time, a manual mode is provided, and parents or teachers can confirm the operation through the APP. The system monitors the implementation effect and provides real-time feedback and adjustments to ensure that the environmental parameters meet the standards. For example, specifically in Xiao Ming's case, the system automatically activates the home humidifier, adjusts the humidity from 35% to 55%, and raises the temperature to 21°C; on the kindergarten side, the teacher manually starts the device after receiving the APP notification. The system then monitors the sensor data to confirm that the humidity is stable at 52% and the temperature is 20.5°C, and generates a report showing that the risk has dropped to medium, effectively reducing the incidence of respiratory diseases. Through this closed-loop management, the system not only optimizes the environmental settings, but also promotes home-school co-education and achieves continuous protection of children's health.

[0075] Next, the present invention further describes the process of assessing the risk of respiratory diseases in young children. This process includes four steps: First, obtaining the child's age data Age (unit: years); second, obtaining the real-time air humidity H (unit: percentage) and air temperature T (unit: degrees Celsius); then, calculating the risk score based on the formula Risk_score = a × (1 / H) + b × |T-25| + c × (1 / Age), where a, b, and c are preset weight coefficients; finally, if Risk_score exceeds the risk threshold R_th (preset value), the risk is determined to be high; otherwise, the risk is low.

[0076] The meaning of each step is as follows: Step 1 obtains age data, as younger children have weaker immune systems and are at higher risk for respiratory illness. Step 2 obtains environmental parameters. Low humidity can easily dry out the respiratory tract, and temperatures outside the comfort zone increase the probability of infection. Step 3 uses a comprehensive formula to quantify risk, with higher scores indicating greater risk. Step 4 determines the risk level based on thresholds to facilitate timely intervention.

[0077] The parameters in the formula are as follows: a represents the humidity impact weight (a positive real number, such as 0.1-1.0, with the optimal value optimized based on data), b represents the temperature deviation weight (same range as a), and c represents the age impact weight (same range as a). H ranges from 0% to 100% (optimal values ​​are higher, such as 50%-70%, to minimize risk), T ranges from common environments, such as -10°C to 40°C (optimal value is 25°C, to minimize deviation), and Age ranges from young children, such as 0-6 years old (optimal value is higher, such as 5-6 years old, to minimize risk). The formula is designed this way because the lower the humidity, the higher the risk (1 / H amplifies the effect of low values), the greater the temperature deviation from 25°C, the higher the risk (|T-25| quantifies the deviation), and the younger the age, the higher the risk (1 / Age amplifies the effect of younger age). This reflects the synergistic effects of environmental and personal factors on respiratory health.

[0078] In one embodiment, based on the home-school co-education system, specifically, the kindergarten collaborates with parents: the system obtains the child's age (Age = 2 years old) from the parent side, and obtains real-time humidity (H = 30%) and temperature (T = 20°C) from the classroom sensor. Preset weights a = 0.5, b = 0.3, c = 0.2, and threshold R_th = 5. Calculate Risk_score = 0.5×(1 / 30) + 0.3×|20-25| + 0.2×(1 / 2) ≈ 0.0167+1.5+0.1 = 1.6167. Since 1.6167 < 5, the system determines it as low risk and notifies parents through the APP to strengthen daily protection, without the need for emergency measures.

[0079] Next, the present invention is described to further assess the risk of respiratory diseases in young children. First, the child's weight data Weight is obtained in kilograms. The data comes from the daily health records entered by parents or teachers in the home-school co-education system. Secondly, the risk score is adjusted based on the weight data, using the formula Risk_score=Risk_score×(1+k_w×(1 / Weight)), where Weight is the weight (usually in the range of 10-30kg, representing the typical weight of young children), k_w is the weight adjustment factor (k_w>0, the range is recommended to be 0.1-1.0, and the optimal value of 0.5 is determined by data analysis to balance sensitivity). The formula means that the lighter the weight, the greater the (1 / Weight) value, resulting in a higher increase in the risk score. This formula is set because young children with light weight have weaker immunity and are susceptible to respiratory diseases. Then, if the adjusted Risk_score>R_th (R_th is the risk threshold, ranging from 0-100, and the optimal value of 70 is set by experts based on epidemiological data), it is determined to be high risk; otherwise, low risk. Finally, a risk level report is generated based on the judgment results and integrated into the home-school co-education platform for parents and teachers to collaborate on preventive measures.

[0080] For example, in one embodiment, the system obtains the weight of a 5-year-old child, Weight = 15 kg (input from the parent APP). The initial Risk_score = 60 (calculated by other health indicators), k_w takes the optimal value of 0.5, and the adjustment calculation is Risk_score = 60 × (1 + 0.5 × (1 / 15)) ≈ 62. R_th is set to 65, so 62 < 65, which is judged to be low risk. Specifically, the system generates a report indicating low risk, recommends increasing nutritional intake in home-school co-education, and shares it with parents and kindergarten teachers through the platform to collaboratively monitor the health of young children.

[0081] Next, we describe the present invention's method for assessing respiratory disease risk in young children. This process includes the following steps: First, obtaining historical disease frequency data (Freq), representing the number of respiratory disease occurrences in young children over the past year; second, updating the risk score based on the formula: Risk_score = Risk_score + d × Freq, where d is the historical disease weight coefficient (d>0); then, if the updated Risk_score > R_th, the risk is determined to be high; otherwise, it is low; finally, the risk level report is correlated with the historical data for analysis.

[0082] Obtaining historical disease frequency data (Freq) means collecting the actual number of respiratory illnesses (such as colds or bronchitis) a child has experienced in the past year to quantify their disease history. For example, in one embodiment, a parent enters the number of colds a child has had in the past year as three through the mobile terminal of the home-school co-education system, and the system automatically records the Freq value.

[0083] When updating the risk score based on the formula, the parameter Freq ranges from non-negative integers (such as 0, 1, 2, ...), and the lower the optimal value, the better, indicating a good health status; d is the weight coefficient (d>0), which usually ranges from 0.1 to 1.0. The optimal value needs to be set through system calibration (such as d=0.5) to balance historical influences; Risk_score is the initial score (which may be obtained from other evaluation modules), and R_th is a preset threshold (such as 3.0). The formula Risk_score=Risk_score+d×Freq means to increase the contribution of historical disease frequency to risk, because frequent disease occurrence indicates higher future risk. Setting this formula can objectively quantify the impact of historical data. Specifically, in one embodiment, the initial Risk_score is 2.0, Freq=4, and d=0.6, then the updated Risk_score=2.0+0.6×4=4.4.

[0084] Determining the risk level means comparing the updated score with the threshold R_th to categorize the risk and facilitate timely intervention. For example, if R_th = 3.0, then 4.4 > 3.0 is considered high risk, and the system generates an alert to notify parents and teachers.

[0085] Correlating risk level reports with historical data allows for analysis to identify trends or patterns, combining historical records (such as disease data from previous years) to optimize assessments. For example, the system can store a child's risk scores over the years, analyze the increased risk during winter, and recommend preventive measures in the report.

[0086] Next, we will further describe the present invention's generation of an environmental control strategy to optimize home environment settings. First, the target humidity H_target and target temperature T_target are determined based on the risk score, Risk_score. The Risk_score represents the child's health risk level and typically ranges from 0 to 100, with higher values ​​indicating greater risk. For example, a Risk_score of 0 indicates the lowest risk, while a Risk_score of 100 indicates the highest risk.

[0087] Secondly, use the formula to calculate \(H_{target}=50 + m\times Risk\_score\) and \(T_{target}=25 + n\times(Risk\_score - R\_th)\). Here, \(m\) and \(n\) are adjustment coefficients, where \(m\lt0\) and \(n\lt0\), and the range can be set from -1 to 0. The optimal values are, for example, \(m = -0.5\) and \(n = -0.2\) (determined by system tuning); \(R\_th\) is the risk threshold, and the range is assumed to be 50 (when \(Risk\_score = R\_th\), \(T_{target}=25\)). The meaning of the formula: Since low humidity exacerbates health risks, the higher the risk (\(Risk\_score\) is large), the target humidity needs to be increased to raise the environmental humidity. However, \(m\lt0\) in the formula causes \(H_{target}\) to decrease as \(Risk\_score\) increases, which aims to indirectly drive an increase in the actual humidity (through subsequent strategies); at the same time, when the risk is higher, the target temperature needs to approach 25°C to reduce the deviation because temperature fluctuations affect the health of young children. \(n\lt0\) in the formula makes \(T_{target}\) approach 25 when \(Risk\_score\) approaches \(R\_th\), reducing extreme values. The formula is set up in this way to dynamically balance the environmental parameters, prioritizing temperature stability and moderately adjusting humidity.

[0088] Next, if the current humidity \(H\lt H_{target}\), the strategy is to increase the humidity; otherwise, maintain or reduce the humidity. This ensures that the environmental humidity is adjusted towards the target value, avoiding the exacerbation of risks due to too low humidity.

[0089] Finally, if the current temperature \(T\lt T_{target}\) or \(T\gt T_{target}\), the strategy is to adjust the temperature to \(T_{target}\). This directly stabilizes the temperature at the target value, reducing health risks.

[0090] Specifically, in one embodiment, assume that the health risk of young children is evaluated based on the home-school cooperation system, \(Risk\_score = 70\) (high risk), \(R\_th\) is set to 50, and \(m\) and \(n\) are taken as -0.5 and -0.2 respectively. Calculate \(H_{target}=50+(-0.5)\times70 = 15\) (low target humidity), \(T_{target}=25+(-0.2)\times(70 - 50)=21\) (target temperature close to 25°C). For example, if the current humidity \(H = 10\) (lower than \(H_{target}\)), the strategy is to increase the humidity; if the current temperature \(T = 23\) (higher than \(T_{target}\)), the strategy is to cool down to 21°C. This example optimizes the home environment and reduces the respiratory risk of young children.

[0091] Next, the generation environment regulation strategy of the present invention is described to further optimize the home environment setting. This strategy includes the following steps: First, obtain the season data Season, where Season represents the current season, encoded as a numerical value (e.g., winter = 1, summer = 2), for identifying seasonal changes; Second, adjust the target humidity and temperature values based on the season, with the formula H_target = H_target × S_h and T_target = T_target × S_t, where H_target is the initial target humidity (unit: %, range 40% - 60%, optimal value 50%), T_target is the initial target temperature (unit: °C, range 18 - 25 °C, optimal value 22 °C), and S_h and S_t are season factors (winter S_h = 2, S_t = 1.1; summer S_h = 0.8, S_t = 0.9). The meaning of this formula is to adjust the target value through the multiplication factor to meet the seasonal requirements. For example, in winter, the factors increase the humidity and temperature to reduce the negative impact of dryness and cold on the respiratory health of young children; Third, if Season is winter and the current humidity H < H_target, the strategy gives priority to increasing the humidity, for example, it is recommended to use a humidification device; Fourth, output the optimized environment regulation strategy, such as specific device operation instructions.

[0092] In a specific example, in an embodiment, it is assumed that it is applied to the comprehensive assessment management method and system for the health status of young children based on home-school cooperation. The current season is winter (Season = 1), and the system obtains the initial H_target = 50% and T_target = 22 °C. Based on the season adjustment: S_h = 2, S_t = 1.1, so H_target is adjusted to 100% (50% × 2), and T_target is adjusted to 24.2 °C (22 °C × 1.1). If the detected actual humidity H = 45% < 100%, the strategy gives priority to increasing the humidity, for example, it outputs to turn on the humidifier to the target humidity of 100%. The reason for this setting is that the dryness and cold in winter are likely to cause skin dryness or colds in young children. The target value is amplified by the factor to strengthen the protection, while the summer factor reduces the target to avoid discomfort caused by high temperature and high humidity. The overall strategy optimizes the home environment and supports the health assessment of young children.

[0093] Next, the implementation of the environmental control strategy of the present invention is further described. The strategy includes the following steps: First, the adjusted air humidity H and air temperature T are monitored in real time, which means that the environmental parameters are continuously collected to ensure that the state after control is tracked. Secondly, the control effect index E is calculated based on the formula E = |H-H_target| / H_target+|T-T_target| / T_target, where E reflects the degree of deviation between the actual and target environments (the smaller the value, the better the effect); the parameter H is the actual humidity (range 40%-60%, optimal 50%), H_target is the target humidity (same range), T is the actual temperature (range 18-22°C, optimal 20°C), and T_target is the target temperature (same range); the formula is normalized by relative error, combining parameters of different units into a dimensionless index to facilitate quantification of the overall deviation. This setting can fairly compare the effects of humidity and temperature. Third, if E>E_th (E_th is a preset effect threshold, for example 0.2), a reassessment of the risk is triggered, meaning that when the deviation is too large, the potential threat to the child's health is re-analyzed. Fourth, dynamically update the control strategy based on the reassessment results, adjusting environmental control measures to optimize conditions. For example, in one embodiment, the system sets H_target = 50% and T_target = 20°C for a kindergarten activity room. Specifically, if the actual H = 55% and T = 22°C, then E = |55-50| / 50+|22-20| / 20 = 0.1+0.1 = 0.2. Assuming E_th = 0.15, a reassessment is triggered, revealing an elevated respiratory risk for the child. This results in an updated strategy, such as lowering the air conditioning temperature to 19°C and increasing the humidifier output.

[0094] Next, the implementation environment control strategy of the present invention is described further.

[0095] Step 1: Introduce a feedback control loop and adjust the coefficients based on the control effect index E: if E>E_th, then m=m×9 and n=n×0.9 (reduce the step size to prevent over-control).

[0096] This step achieves adaptive control by dynamically adjusting coefficients: when the control effect index E exceeds the threshold E_th, m is increased to accelerate the response, while n is decreased to reduce the step size and avoid oscillation of environmental parameters. The parameter E represents the control effect (range [0, 1], with 1 being optimal), E_th is the preset threshold (range [0.7, 0.9], with an optimal value of 0.8), and m and n are the control coefficients (the initial values ​​are set by the system and have an unlimited range; the optimal values ​​require tuning). The reason for this formula setting is that E > E_th indicates an overly strong effect, so increasing m improves efficiency and decreasing n suppresses the step size to ensure stable convergence.

[0097] For example, in kindergarten temperature control, if E calculated based on the child's health assessment (such as comfort score) is 0.85>E_th(0.8), the system will increase m from 1.0 to 9.0 and decrease n from 1.0 to 0.9 to prevent sudden temperature changes from affecting the child's health.

[0098] Step 2: Recalculate H_target and T_target based on the updated m and n.

[0099] This step regenerates the target values ​​using the new coefficients: H_target represents the child's health target (e.g., immunity index) and T_target represents the environmental target (e.g., ideal temperature). The formulas H_target = f(m,n,input data) and T_target = g(m,n,input data) ensure that the target values ​​adapt to real-time feedback.

[0100] Specifically, in one embodiment, the system uses the updated m=9.0 and n=0.9, combined with the children's body temperature data, to recalculate T_target to 24° C. (originally 22° C.) to optimize the classroom environment.

[0101] Step 3: If E≤E_th for three consecutive times, maintain the current strategy; otherwise, continue to adjust.

[0102] This step achieves strategy stability through continuous evaluation: when E does not exceed the threshold for three consecutive times, it indicates that the regulation effect is stable and adjustments are suspended; otherwise, the optimization is cyclically optimized to avoid ineffective intervention.

[0103] For example, in the humidity management of young children, if E based on respiratory health index is ≤ E_th(0.8) for three consecutive times, the system maintains the current humidification strategy; otherwise, the adjustment cycle is restarted.

[0104] Step 4: Store feedback data for historical analysis.

[0105] This step stores data such as E, m, and n to support long-term trend analysis and optimize future strategies.

[0106] In one embodiment, the system archives temperature and health data to assist in home-based parenting report generation, such as identifying seasonal disease patterns.

[0107] (Word count: 498)

[0108] Next, the present invention is further described for assessing the risk of respiratory diseases in young children.

[0109] First, the steps include: (1) using a machine learning model to predict the risk score, with the input features being H (height, in cm), T (body temperature, in °C), Age (in years), and Weight (in kg); and (2) optimizing the model weights based on the formula:

[0110] Δw = η × (Risk_score_actual - Risk_score_predicted) × input features;

[0111] (3) If the prediction error |Risk_score_actual-Risk_score_predicted|>error threshold (preset value), retrain the model; (4) Output the optimized risk score for generating health management strategies.

[0112] The meaning of each step is as follows: Step (1) calculates the risk score of respiratory diseases in young children based on the input features through a model (such as linear regression), ranging from 0 to 1 (0 represents low risk and 1 represents high risk). The formula in step (2) is used to update the weights and reduce the prediction error through gradient descent; where η (learning rate) controls the update step size, ranging from 0.001 to 0.1 (optimal value 0.01), risk_score_actual comes from real medical data, risk_score_predicted is the model output, and the input features such as H or T are standardized. The formula setting is designed to optimize the model efficiently because it directly uses the error gradient to adjust the weights and improve accuracy. Step (3) triggers retraining when the absolute value of the error exceeds a threshold (such as 0.05) to ensure model adaptability. Step (4) inputs the final risk score into the strategy module to generate personalized recommendations.

[0113] In one embodiment, the home-school co-education system collects data for a 3-year-old child: H = 95 cm, T = 36.8°C, Age = 3, Weight = 15 kg. The model predicts Risk_score_predicted = 0.4 (low risk), but the actual diagnosis is Risk_score_actual = 0.6 (due to a mild cold). The error |0.6 - 0.4| = 0.2 is greater than the threshold of 0.1, so the system retrains the model: using the formula Δw to update the weights (for example, η = 0.01), the optimized predicted score is adjusted to 0.55. The output is used to generate policies, such as notifying parents to increase ventilation or arrange health checks within the kindergarten.

[0114] Next, the use of the machine learning model to generate environmental regulation strategies to further optimize the home environment settings of the present invention is described. The steps include: (1) Designating the model as a neural network, where the input layer includes H (humidity), T (temperature), Age (age), Weight (weight), Freq (frequency), and the output layer is Risk_score (risk score); (2) Judging the activation of the hidden layer based on the standardized values of the input features: If the standardized value > 5, the ReLU activation function is used, otherwise the Sigmoid function is used; (3) Training the model until the loss function L = (1 / N) × Σ(Risk_score_actual - Risk_score_predicted)^2 < L_th (loss threshold), where N is the number of samples; (4) Deploying the trained model for real-time risk assessment; (5) Adding a user confirmation step: Pushing the regulation strategy to parents or teachers; (6) Adjusting the strategy based on user feedback: If agreed, implement it. If modified, update H_target = H_target × F_h and T_target = T_target × F_t, where F_h and F_t are feedback factors (F_h = 95 means a slight decrease in humidity, F_t = 1.05 means a slight increase in temperature); (7) If no feedback is received, delay the implementation of the strategy until timeout; (8) Recording user interaction data for subsequent strategy optimization.

[0115] The meaning of each step is as follows: (1) Establishing a neural network model with environmental parameters and child characteristics as inputs to predict health risks; (2) Selecting an activation function according to the feature values. ReLU is used for high-value features to handle non-linearity, and Sigmoid is used for low-value features to smooth the output; (In the training process, the mean squared error loss function is used to minimize the difference between the predicted and actual risk scores and ensure the accuracy of the model; (4) Applying the model to real-time monitoring to dynamically evaluate risks; (5) Pushing the strategy to users to ensure transparent decision-making; (6) Fine-tuning the target parameters based on feedback. F_h = 95 (range 90 - 100, optimal 95) means a 5% decrease in humidity, and F_t = 1.05 (range 1.0 - 1.1, optimal 1.05) means a 5% increase in temperature. The formula settings allow for gradual adjustment to avoid sudden changes; (7) The delay mechanism prevents misoperations; (8) Data recording is used for iterative improvement of the model.

[0116] Formula Explanation: In the loss function L, Risk_score_actual and Risk_score_predicted range from 0 to 100 (optimal value close to 0), N is a positive integer, and the formula calculates the sum of squared prediction errors divided by N, representing the mean error. This setting, because mean squared error is a regression criterion, penalizes large errors and stabilizes training. In the update formulas H_target and T_target, F_h and F_t are based on user preferences (F_h = 95 represents 95% of the original value), with a range of 90-100 for F_h and 1.0-1.1 for F_t. The optimal value is determined by feedback, and the formula is fine-tuned through multiplication because linear changes are simple and easy to control.

[0117] For example, in one embodiment, for a kindergarten child (Age = 4 years old, Weight = 18 kg), the system inputs H = 60%, T = 22°C, and Freq = 3 times / day, and the model predicts Risk_score = 30 (indicating medium risk). After deployment, the strategy recommends lowering the humidity to 55%. After the parents are pushed and feedback is provided, the update H_target = 55×0.95≈52.25%, T_target = 22×1.05≈23.1°C is updated. If there is no feedback within the timeout, the implementation is delayed; the data records are used to optimize subsequent strategies, such as adjusting the model weights. Specifically, this process optimizes the home environment and reduces the risk of colds in young children. (Word count: 498)

[0118] The comprehensive assessment and management method and system of the health status of young children based on home-school co-education of the present invention include: first, through the integrated sensor network and user input interface, the health status data of young children and the health factor data of the home environment are acquired in real time. The health status data of young children covers physiological indicators (such as body temperature, respiratory rate, cough symptoms, sleep quality and diet records), as well as behavioral data (such as activity level and emotional state), which are collected through wearable devices (such as smart bracelets) or parent reporting APPs. At the same time, the environmental health factor data include air humidity and temperature, which are automatically monitored by temperature and humidity sensors (such as smart thermometers and hygrometers) deployed in the home environment and uploaded to the cloud database in real time. This step solves the problem of the comprehensiveness of the initial data collection and ensures that subsequent analysis is based on multi-source heterogeneous data.

[0119] Secondly, based on the acquired data, a machine learning model (such as a random forest or neural network algorithm) is used to assess the risk of respiratory diseases in young children. The model is trained on a historical data set that associates environmental factors (such as air humidity below 40% or temperature below 18°C) with the incidence of respiratory diseases in young children (such as colds or asthma). Specifically, the system analyzes whether the current humidity and temperature exceed the preset thresholds (for example, humidity range 40%-60%, temperature range 20-24°C), and combines the health status of the child (such as abnormal body temperature or cough frequency) to calculate a risk score (such as low, medium, and high levels). This step solves the problem of dynamic quantification of the association between environmental factors and diseases, avoiding subjective judgment.

[0120] Then, based on the risk assessment results, an environmental control strategy is generated to optimize the home environment settings. The system has a built-in policy engine that matches the preset rule base according to the risk level: for example, if the assessment is high risk (humidity <40% and the child has coughing symptoms), the generated strategy includes increasing the humidity to 50% and raising the temperature to 22°C; if it is low risk, it is only recommended to maintain the current settings or fine-tune. The strategy also takes into account the characteristics of home-school co-education, pushes personalized suggestions to parents through the APP (such as turning on a humidifier or adjusting the air-conditioning mode), and integrates smart home control protocols (such as connecting devices through the IoT interface). This step solves the real-time and adaptability of the control, ensuring that the environment is optimized for the susceptible factors of young children.

[0121] Finally, the environmental control strategy is implemented, and the air humidity and temperature in the home environment are adjusted through an automated control system. The system connects to smart devices (such as smart humidifiers and constant temperature air conditioners) and automatically performs operations according to policy instructions (such as starting a humidifier to increase humidity or adjusting the air conditioner temperature). At the same time, the system monitors the implementation effect, verifies whether the adjustment meets the standards through feedback loops (such as sensors sending back new data in real time), and iteratively optimizes the strategy. If manual intervention is required, parents can confirm or adjust it through the APP. The overall process solves the problem of young children being susceptible to respiratory diseases: through closed-loop management, environmental factors (such as humidity and temperature) are actively regulated to a healthy range, reducing triggering factors such as dryness or coldness, thereby reducing the risk of disease and improving the health level of young children. This method and system emphasize home-school collaboration to achieve preventive health management.

[0122] The specific implementation methods described above further illustrate the purpose, technical solutions and beneficial effects of the embodiments of the present disclosure. It should be understood that the above description is only a specific implementation method of the embodiments of the present disclosure and is not intended to limit the scope of protection of the embodiments of the present disclosure. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the embodiments of the present disclosure should be included in the scope of protection of the embodiments of the present disclosure.

Claims

1. A comprehensive evaluation and management method and system for children's health status based on home-school co-education, characterized by: include: Acquiring child health status data and home environment health factor data, wherein the environmental health factors include air humidity and temperature; Assessing the risk of respiratory diseases in young children based on the health status data and environmental health factor data of the young children; Based on the respiratory disease risk assessment results, generate environmental control strategies to optimize home environment settings; Implement the environmental control strategy to adjust the air humidity and temperature in the home environment.

2. The comprehensive evaluation and management method and system for children's health status based on home-school co-education according to claim 1 is characterized in that: The assessment of the risk of respiratory diseases in young children further includes: Get the child's age data Age, where Age represents the child's age (unit: years); Get real-time air humidity H and air temperature T, where H represents air humidity (unit: percentage) and T represents air temperature (unit: degrees Celsius); The risk score is calculated based on the following formula: Risk_score = a × (1 / H) + b × |T25| + c × (1 / Age), where a, b, and c are preset weight coefficients (a represents the humidity impact weight, b represents the temperature deviation weight, and c represents the age impact weight). This formula quantifies risk: the lower the humidity, the greater the temperature deviation from 25°C, or the younger the age, the higher the risk. If Risk_score>risk threshold R_th (R_th is a preset threshold), it is judged as high risk; otherwise it is judged as low risk.

3. The comprehensive evaluation and management method and system for children's health status based on home-school co-education according to claim 2 is characterized in that: The assessment of the risk of respiratory diseases in young children further includes: Get the child's weight data Weight, where Weight represents the child's weight (unit: kilogram); Adjust the risk score based on weight data: Risk_score = Risk_score × (1 + k_w × (1 / Weight)), where k_w is the weight adjustment factor (k_w>0). This adjustment means that the lighter the weight, the greater the risk score increase; If the adjusted Risk_score>R_th, it is judged as high risk; otherwise, it is judged as low risk; Generate a risk level report based on the determination results.

4. The comprehensive evaluation and management method and system for children's health status based on home-school co-education according to claim 3 is characterized in that: The assessment of the risk of respiratory diseases in young children further includes: Obtain historical disease frequency data Freq, where Freq represents the number of respiratory diseases that occurred in young children in the past year; Update the risk score based on the following formula: Risk_score = Risk_score + d × Freq, where d is the historical disease weight coefficient (d>0). This update means that the higher the historical disease frequency, the higher the risk score; If the updated Risk_score>R_th, it is judged as high risk; otherwise, it is judged as low risk; Correlate and analyze risk level reports with historical data.

5. The comprehensive evaluation and management method and system for children's health status based on home-school co-education according to claim 1 is characterized in that: Generating an environmental control strategy to optimize home environment settings further includes: Determine the target humidity H_target and target temperature T_target based on the risk score Risk_score; Calculate \(H_{target}=50 + m\times Risk\_score\) and \(T_{target}=25 + n\times(Risk\_score - R_{th})\) based on the following formula, where \(m\) and \(n\) are regulation coefficients (\(m\lt0\), \(n\lt0\)). The meaning of this formula is that the higher the risk, the higher the target humidity needs to be increased (because low humidity exacerbates the risk), and the target temperature needs to approach \(25^{\circ}C\) (to reduce the deviation). If the current humidity \(H\lt H_{target}\), the strategy is to increase the humidity; otherwise, maintain or reduce the humidity. If the current temperature \(T\lt T_{target}\) or \(T\gt T_{target}\), the strategy is to adjust the temperature to \(T_{target}\).

6. The comprehensive evaluation and management method and system for children's health status based on home-school co-education according to claim 5 is characterized in that: The generation of the environmental regulation strategy for optimizing the home environment settings further includes: Obtain the season data Season, where Season represents the current season (encoded as a numerical value, e.g., winter = 1, summer = 2). Adjust the target values based on the season: \(H_{target}=H_{target}\times S_h\) and \(T_{target}=T_{target}\times S_t\), where \(S_h\) and \(S_t\) are season factors (in winter, \(S_h = 2\), \(S_t = 1.1\); in summer, \(S_h = 0.8\), \(S_t = 0.9\)). This adjustment means that higher humidity and temperature are required in winter to reduce the impact of dryness and cold. If Season is winter and \(H\lt H_{target}\), the strategy gives priority to increasing the humidity. Output the optimized environmental regulation strategy.

7. The comprehensive evaluation and management method and system for children's health status based on home-school co-education according to claim 1 is characterized in that: The implementation of the environmental regulation strategy further includes: Real-time monitor the adjusted air humidity \(H\) and air temperature \(T\). Calculate the regulation effect index \(E=\frac{|H - H_{target}|}{H_{target}}+\frac{|T - T_{target}|}{T_{target}}\) based on the following formula, where \(E\) represents the degree of deviation between the actual and target environments (the smaller the value, the better the effect). If \(E\gt E_{th}\) (\(E_{th}\) is the preset effect threshold), trigger a re-evaluation of the risk. Dynamically update the regulation strategy according to the re-evaluation results.

8. The comprehensive evaluation and management method and system for children's health status based on home-school co-education according to claim 7 is characterized in that: The implementation of the environmental regulation strategy further includes: Introduce a feedback control loop to adjust the coefficients based on the regulation effect index \(E\): if \(E\gt E_{th}\), then \(m = m\times9\) and \(n = n\times0.9\) (reduce the step size to prevent over-regulation). Recalculate \(H_{target}\) and \(T_{target}\) based on the updated \(m\) and \(n\). If \(E\leq E_{th}\) for three consecutive times, maintain the current strategy; otherwise, continue to adjust. Store the feedback data for historical analysis.

9. The comprehensive evaluation and management method and system for children's health status based on home-school co-education according to claim 1 is characterized in that: The evaluation of the risk of childhood respiratory diseases further includes: Use a machine learning model to predict the risk score, and the inputs of the model include \(H\), \(T\), Age, Weight. Optimize the model weights based on the following formula: the weight update \(\Delta w=\eta\times(Risk\_score_{actual}-Risk\_score_{predicted})\times input\_feature\), where \(\eta\) is the learning rate, \(Risk\_score_{actual}\) is the actual risk score, and \(Risk\_score_{predicted}\) is the model prediction value. The meaning of this formula is to reduce the prediction error through gradient descent. If the prediction error |Risk_score_actual - Risk_score_predicted| > error threshold, retrain the model; Output the optimized risk score for generating the strategy.

10. The comprehensive evaluation and management method and system for children's health status based on home-school co-education according to claim 9 is characterized in that: The use of the machine learning model to generate the environmental regulation strategy to optimize the home environment settings further includes: Specify the model as a neural network, the input layer includes H, T, Age, Weight, Freq, and the output layer is Risk_score; Judge to activate the hidden layer based on the following conditions: if the standardized value of the input feature > 5, use the ReLU activation function; otherwise use Sigmoid; Train the model until the loss function L = (1 / N) × Σ(Risk_score_actual - Risk_score_predicted)^2 < L_th (L_th is the loss threshold), where N is the number of samples; Deploy the trained model for real-time risk assessment; Add a user confirmation step: push the regulation strategy to parents or teachers; Adjust the strategy based on user feedback: if the feedback is consent, implement the strategy; if the feedback is modification, update H_target = H_target × F_h and T_target = T_target × F_t based on the following formula, where F_h and F_t are feedback factors (F_h = 95 means slightly decreasing humidity, F_t = 1.05 means slightly increasing temperature); If no feedback is received, delay implementing the strategy until timeout; Record user interaction data for subsequent strategy optimization.

Images