A multi-data fusion health risk early warning method and system

Abstract

The application provides a multi-data fusion health risk early warning method and system, relates to the field of health risk early warning, and comprises the following steps: defining a plurality of load categories and a plurality of physiological resource accounts corresponding to the load categories in advance; acquiring external data corresponding to a specific load category in the load categories, and quantifying the specific load category based on the external data to obtain a load quantification value; according to the load quantification value, performing consumption calculation on an account state value of a physiological resource account corresponding to the specific load category to update the account state value; when a preset evaluation condition is met, based on a comparison result of the updated account state values of the plurality of physiological resource accounts, an evaluation output indicating a recovery priority is generated, which has the advantages that multi-source data can be comprehensively considered, the influence of different loads on physiological resources can be quantified, and a targeted recovery priority suggestion can be generated based on the resource consumption state, thereby improving the accuracy and effectiveness of health risk early warning and recovery guidance.

Description

Technical Field

[0001] This application relates to the field of health risk early warning, and more specifically, to a multi-data fusion health risk early warning method and system. Background Technology

[0002] In the field of sports health monitoring, especially in outdoor sports scenarios, accurate assessment of athletes' physical condition and risk warning are of great significance. Existing sports monitoring systems typically rely on collecting athletes' physiological readings, such as heart rate, and combining them with exercise intensity data (such as speed, power, and altitude changes) to assess physical load and energy consumption. However, outdoor environmental conditions are complex and variable, such as temperature, humidity, and altitude. These environmental factors can directly or indirectly affect athletes' physiological responses and may interfere with or deviate from the readings of the sensing units.

[0003] Specifically, an athlete's physiological readings are often the result of the combined effects of their own exercise and external environmental factors. For example, in a high-temperature environment, the body increases heart rate and sweating to dissipate heat, which means that at the same exercise intensity, the heart rate may be higher than in a cooler environment.

[0004] Furthermore, when environmental conditions change rapidly, an athlete's physiological state undergoes an adaptive adjustment process, manifesting as fluctuations in physiological readings. If the system fails to distinguish these environmental adaptive fluctuations from genuine bodily dysfunctions, it may trigger false alarms, interfering with the athlete's normal activities. Simultaneously, extreme environmental conditions may also affect the performance of the sensing unit itself and the accuracy of the data, further increasing the complexity of data interpretation.

[0005] To address the aforementioned issues, existing technologies urgently need improvement. Summary of the Invention

[0006] The purpose of this application is to provide a multi-data fusion health risk early warning method and system, which has the advantages of comprehensively considering multi-source data, quantifying the impact of different loads on physiological resources, and generating targeted recovery priority suggestions based on resource consumption status, thereby improving the accuracy and effectiveness of health risk early warning and recovery guidance.

[0007] This application provides a multi-data fusion health risk early warning method, the technical solution of which is as follows:

[0008] include:

[0009] Multiple load categories and corresponding physiological resource accounts are predefined;

[0010] Obtain external data corresponding to a specific load category in the load category, and quantize the specific load category based on the external data to obtain the load quantization value;

[0011] Based on the load quantification value, the account status value of the physiological resource account corresponding to a specific load category is consumed and the account status value is updated.

[0012] When the preset evaluation conditions are met, an evaluation output indicating the recovery priority is generated based on the comparison results of the updated account status values ​​of multiple physiological resource accounts.

[0013] The above approach can comprehensively consider multi-source data, quantify the impact of different workloads on physiological resources, and generate targeted recovery priority suggestions based on resource consumption status, thereby improving the accuracy and effectiveness of health risk warning and recovery guidance.

[0014] Furthermore, this application also proposes a multi-data fusion health risk early warning system, the technical solution of which is as follows:

[0015] The system for implementing the above-mentioned multi-data fusion health risk early warning method includes:

[0016] Define a storage module for pre-defining multiple load categories and multiple physiological resource accounts corresponding to the load categories;

[0017] The data acquisition and quantization module is used to acquire external data corresponding to a specific load category in the load category, and to quantize the specific load category based on the external data to obtain the load quantization value;

[0018] The consumption calculation and update module is used to calculate the consumption of the account status value of the physiological resource account corresponding to a specific load category based on the load quantification value, so as to update the account status value.

[0019] The evaluation output module is used to generate an evaluation output indicating the recovery priority based on the comparison results of the updated account status values ​​of multiple physiological resource accounts when the preset evaluation conditions are met.

[0020] As can be seen from the above, the multi-data fusion health risk early warning method and system provided in this application solves the problems in the prior art of effectively handling the complex interactive effects of multi-source data, distinguishing the dominant sources of physiological stress, and generating accurate recovery strategies by means of multi-data fusion, load quantification, resource account consumption assessment, and priority generation. It has the advantages of being able to comprehensively consider multi-source data, quantify the impact of different loads on physiological resources, and generate targeted recovery priority suggestions based on resource consumption status, thereby improving the accuracy and effectiveness of health risk early warning and recovery guidance. Attached Figure Description

[0021] Figure 1 This is a schematic diagram of a multi-data fusion health risk early warning method provided in this application.

[0022] Figure 2 A schematic diagram of a multi-data fusion health risk early warning system provided in this application. Detailed Implementation

[0023] The technical solutions of this application will now be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are merely some embodiments of this application, and not all embodiments. The components of this application described and shown in the accompanying drawings can generally be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of this application provided in the accompanying drawings is not intended to limit the scope of the claimed application, but merely to illustrate selected embodiments of this application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without inventive effort are within the scope of protection of this application.

[0024] This application initially attempts to establish a direct mapping relationship between physiological indicators and single load factors. However, this method struggles to effectively isolate the cumulative effects of environmental factors on physiological indicators and cannot handle the problem of environmental interference with sensor data. Therefore, this application further explores the need for a method that can comprehensively consider the impact of multiple load categories on the body. This requires quantifying different load categories and linking them to the consumption of different physiological resources within the body. By dynamically tracking the account status of these physiological resources and comparing their status when assessment is needed, the overall stress level of the body and the degree of consumption of different resources can be more accurately determined, thereby generating targeted recovery recommendations.

[0025] Reference Figure 1 The diagram illustrates an embodiment of a multi-data fusion health risk early warning method according to an embodiment of the present invention, which may specifically include the following steps:

[0026] S101, predefine multiple load categories and multiple physiological resource accounts corresponding to the load categories;

[0027] S102, acquire external data corresponding to a specific load category in the load category, and quantize the specific load category based on the external data to obtain the load quantization value;

[0028] S103, Based on the load quantification value, calculate the consumption of the account status value of the physiological resource account corresponding to the specific load category, and update the account status value.

[0029] S104, when the preset evaluation conditions are met, an evaluation output indicating the recovery priority is generated based on the comparison results of the updated account status values ​​of multiple physiological resource accounts.

[0030] Among these, multiple load categories refer to a collection of various types of external or internal factors that affect a user's physical condition. These can be implemented using a pre-defined classification system, such as exercise load, environmental load, and psychological load. They are primarily used for structured management and analysis of factors influencing a user's health. Multiple physiological resource accounts refer to a set of virtual accounts corresponding to the aforementioned load categories, representing the user's internal physiological resource reserves or states. These can be implemented using data structures or models, such as energy accounts, water accounts, electrolyte accounts, and core temperature accounts. They are primarily used to quantify and track the impact of different loads on internal resources. External data refers to information obtained from outside the user's body that is related to a specific load category. This can be obtained through sensor acquisition, user input, or third-party data sources, such as ambient temperature, humidity, altitude, exercise trajectory, heart rate, and body surface temperature. It is primarily used to provide input for load quantification. The quantified load value refers to the numerical representation of the degree of influence of external data on a specific load category. This value can be obtained using a computational model. The system can be implemented using methods such as metabolic equivalents, heat stress index, and exercise intensity scores. These methods primarily convert different types of loads into calculable consumption data. Account status values ​​refer to the current state or remaining amount of the physiological resource account, which can be expressed numerically, as a percentage, or in levels. They primarily reflect the real-time status of the user's corresponding physiological resources. Consumption calculation refers to the process of calculating the impact of specific load categories on the corresponding physiological resource account status values ​​based on the load quantification value and then deducting or adjusting accordingly. This can be implemented using preset consumption models or algorithms and is primarily used to simulate the consumption process of bodily resources by loads. Recovery priority refers to determining the urgency or importance of recovering different physiological resources based on the comparison results of physiological resource account status values. This can be expressed using sorting or grading methods and is primarily used to guide users in targeted recovery. Assessment output refers to user health risk warnings or recovery suggestions generated based on recovery priorities. These can be presented in the form of text, charts, or voice prompts and are primarily used to provide users with actionable guidance.

[0031] The core innovation of this application lies in establishing a correspondence between multiple load categories and multiple physiological resource accounts, quantifying the load based on external data, dynamically updating the status values ​​of physiological resource accounts, and finally generating an assessment output indicating the priority of recovery based on the comparison results of account status values. This solves the problem in the background technology of difficulty in distinguishing the dominant source of physiological stress and the inability to provide personalized and prioritized recovery strategies, achieving the effect of more accurate health risk warning and more effective recovery guidance.

[0032] The solution proposed in this application achieves the aforementioned objectives by constructing a dynamic assessment framework that links external loads with internal physiological resource consumption. First, by pre-defining different types of load categories and their corresponding physiological resource accounts, a foundation is laid for subsequent quantification and assessment. Then, external data related to specific load categories is acquired, and these data are used to quantify the load category, transforming complex external influences into unified load quantification values. This allows loads from different sources to be incorporated into the same assessment system. It is precisely because of these quantified load values ​​that the account status values ​​of the physiological resource accounts corresponding to specific load categories can be calculated and updated based on these values, thereby dynamically reflecting changes in the body's resource consumption and reserves under different loads. When preset assessment conditions are met, the system compares the updated account status values ​​of multiple physiological resource accounts. By comparing the status of different accounts, it can identify which physiological resources are more severely depleted or in a more dangerous state, thereby generating an assessment output indicating recovery priorities to guide users to prioritize replenishing or restoring critical physiological resources.

[0033] In some preferred embodiments, this application is implemented as follows: Load categories such as exercise load, environmental load, and psychological load can be predefined, along with physiological resource accounts such as energy accounts, water accounts, electrolyte accounts, and core temperature accounts. A correspondence between them is established; for example, exercise load primarily consumes energy and electrolytes, while environmental load primarily affects water and core temperature. External data is acquired, such as physiological indicators like heart rate, cadence, body surface temperature, and sweat volume obtained through wearable devices, and environmental information like ambient temperature, humidity, and air pressure obtained through environmental sensors. Based on this external data, specific load categories are quantified. For example, the quantified value of exercise load (e.g., metabolic equivalent) is calculated based on heart rate and cadence, and the quantified value of environmental load (e.g., wet-bulb thermometer) is calculated based on ambient temperature, humidity, and body surface temperature. Based on these quantified load values, the account status values ​​of the corresponding physiological resource accounts are calculated for consumption. For example, energy consumption is calculated based on the quantified exercise load value, and the energy account status value is updated; water loss is calculated based on the quantified environmental load value, and the water account status value is updated. When exercise ends or the account status value falls below a certain threshold, the preset evaluation conditions are met. At this time, the current status values ​​of accounts such as energy account, water account, electrolyte account, and core temperature account are compared. For example, if the water account status value is found to be the lowest, followed by the energy account, an evaluation output indicating the recovery priority is generated, prompting the user to replenish water first, and then replenish energy.

[0034] The aforementioned technical solutions effectively differentiate the impact of different factors, such as exercise load and environmental load, on the body's physiological state, avoiding potential biases that may arise from relying solely on a single physiological indicator. By dynamically tracking the consumption of different physiological resources, the system can more accurately assess a user's current health risk status. Based on a comparison of the status of multiple physiological resource accounts, it can generate recovery suggestions with clear priorities, guiding users to conduct more targeted and effective recovery, thus improving the accuracy of health risk warnings and the effectiveness of recovery guidance.

[0035] In some of the embodiments described above in this application, a method is proposed to calculate the consumption of the account status value of the physiological resource account corresponding to a specific load category based on the load quantification value, so as to update the account status value. Specifically, the consumption calculation based on the load quantification value can be achieved by consulting a preset consumption model or lookup table to directly map the quantified load value to a resource consumption amount. For example, if the exercise intensity quantification value is moderate, the preset moderate consumption value of the physical fitness account is directly deducted. This allows for a quick estimation of resource consumption based on external data. However, in its implementation, relying solely on the load quantification value to update the account status value ignores the impact of the user's actual physiological state on resource consumption, which may lead to a discrepancy between the updated account status value and the user's actual consumption, thereby affecting the accuracy of health risk warnings.

[0036] In response, this application further proposes a step of calculating the consumption of the account status value of the physiological resource account corresponding to a specific load category based on the load quantification value, in order to update the account status value, including:

[0037] Obtain physiological indicators that reflect the user's consumption status;

[0038] Determine whether the load quantification value and physiological indicators meet the first preset threshold condition;

[0039] If the first preset threshold condition is met, an adjusted resource consumption value is determined based on physiological indicators, and the account status value of the physiological resource account is calculated based on the adjusted resource consumption value.

[0040] If the first preset threshold condition is not met, the consumption of the account status value of the physiological resource account corresponding to the specific load category is calculated based on the load quantification value.

[0041] Among them, physiological indicators reflecting the user's consumption status refer to biological parameters that can directly or indirectly reflect the body's current energy metabolism, fatigue level, stress level, or recovery status. These can be achieved using physiological measurement data such as heart rate variability, blood lactate concentration, body temperature, skin conductance, respiratory rate, and electromyography. The first preset threshold condition refers to the criterion used to determine whether there is a significant difference or inconsistency between the load quantification value and the physiological indicators. This can be achieved by comparing whether the relative or absolute deviation between the load quantification value and the consumption level indicated by the physiological indicators exceeds a threshold, or by judging whether the trends of the two are consistent within a preset time window. The adjusted resource consumption value refers to the resource consumption amount that is more closely related to the user's actual consumption situation, obtained by recalculating or correcting based on the physiological indicators when there is a significant difference between the load quantification value and the physiological indicators. This can be achieved by consulting the correction coefficient table based on the physiological indicators, calculating it through a prediction model trained based on the physiological indicators, or weighting the physiological indicators and the load quantification value.

[0042] This application's solution modifies the resource consumption calculation process based on load quantification by introducing physiological indicators that reflect the user's actual consumption status. First, physiological indicators are acquired, and then the consistency between the load quantification value and the physiological indicators is assessed. If there is a significant difference, it indicates that relying solely on the load quantification value may be inaccurate. In this case, the solution determines an adjusted resource consumption value based on the physiological indicators and uses this adjusted value to update the account status value, thus more accurately reflecting the user's actual consumption. If the two are largely consistent, the calculation method based on load quantification value continues. This method of dynamically adjusting the consumption calculation logic according to the actual physiological state allows the physiological resource account status value to more realistically reflect the user's physical resource reserves. By integrating this more accurate consumption calculation into a multi-data fusion health risk early warning framework—that is, in the overall process of defining load categories and accounts, acquiring external data to quantify the load, performing consumption calculations to update the account status value, and generating assessment output based on account status value comparisons—the accuracy of the account status value is improved. This, in turn, makes subsequent health risk assessments and recovery priority judgments based on account status values ​​more reliable and targeted.

[0043] In some preferred embodiments, a specific load category can be exercise load, and the corresponding physiological resource account can be a fitness account. External data is quantified to obtain a load quantification value; for example, by analyzing data such as exercise trajectory, speed, and incline, it is quantified into an exercise intensity index. Physiological indicators reflecting the user's consumption status are acquired; for example, heart rate variability (HRV) data is obtained through wearable devices. A decrease in HRV typically reflects physical fatigue and increased consumption. It is determined whether the exercise intensity index and HRV meet a first preset threshold condition. For example, the preset condition could be that the relative difference between the consumption level indicated by the exercise intensity index and the consumption level indicated by the HRV decrease exceeds 20%. The exercise intensity index can be mapped to an estimated consumption level through a lookup table or model, and the HRV decrease can also be mapped to another estimated consumption level. If the condition is met, for example, the exercise intensity index is not high but the HRV decreases sharply, indicating that the user may be in a state of fatigue and the actual consumption is greater than indicated by the exercise intensity index, then an adjusted resource consumption value is determined based on the HRV. For example, based on the HRV decrease, a correction table is consulted to obtain a correction coefficient, and the initial consumption value corresponding to the exercise intensity index is multiplied by this correction coefficient to obtain the adjusted resource consumption value. Then, update the fitness account status value using this adjusted resource consumption value. If the conditions are not met, for example, if the exercise intensity index is high and the HRV drops sharply, and both are consistent, then update the fitness account status value directly based on the initial consumption value corresponding to the exercise intensity index.

[0044] The above technical solution can more accurately reflect the user's actual physiological resource consumption, avoid the consumption estimation bias that may be caused by relying solely on external load data, and enable the status value of the physiological resource account to more realistically reflect the user's physical resource reserve level, thereby improving the accuracy and reliability of health risk assessment and recovery priority judgment based on account status value.

[0045] In some embodiments described above, an assessment output indicating recovery priority is proposed based on a comparison of the updated account status values ​​of multiple physiological resource accounts. Specifically, this assessment can be achieved by directly comparing the current status values ​​of each account to rank them, thus quickly obtaining a preliminary recovery order. However, in practice, a simple comparison based solely on account status values ​​may not accurately reflect the body's true recovery needs. For example, when the account status values ​​of multiple physiological resource accounts are all low, a simple comparison may fail to distinguish which accounts are declining faster and require priority recovery.

[0046] In response, this application further proposes a step for generating an assessment output indicating recovery priority based on a comparison of updated account status values ​​of multiple physiological resource accounts, including:

[0047] Determine whether there are at least two account status values ​​among the updated account status values ​​of multiple physiological resource accounts that meet a preset numerical difference threshold condition.

[0048] If the numerical difference threshold condition is met, the historical account status values ​​of at least two physiological resource accounts that meet the numerical difference threshold are obtained within a preset time period. Based on the historical account status values, the rate of change of the status value of each account is determined. Based on the comparison results of the rate of change of the status value, the recovery priority is determined and an evaluation output is generated.

[0049] If the numerical difference threshold condition is not met, the recovery priority is determined based on the comparison results of the updated account status values ​​of multiple physiological resource accounts, and an evaluation output is generated.

[0050] The preset numerical difference threshold condition refers to the numerical limit used to determine whether the difference between at least two account status values ​​reaches a preset level. Specifically, it can be a fixed value, such as 10 units, or a relative proportion, such as 20% of the difference between the maximum and minimum values. Its purpose is to distinguish between significant and insignificant differences in account status values, allowing for different priority determination strategies. The preset time period refers to the time range used to acquire historical account status values, specifically one hour, one day, or one week, etc., aiming to provide a sufficient time span to analyze the changing trends of account status values. Historical account status values ​​refer to the sequence of physiological resource account status values ​​recorded within the preset time period. Specifically, it can be timestamp data points associated with each account read from a database or storage medium, aiming to provide a data basis for calculating the rate of change of status values. The rate of change of status values ​​refers to the rate of change of physiological resource account status values ​​within the preset time period. The rate or trend of change during the period can be determined by calculating the slope, average rate of change, or derivative of the fitted curve of the historical account status value sequence. Its purpose is to reflect the speed of account status deterioration. Determining recovery priority based on the comparison of status value change rates means ranking the recovery order according to the magnitude of the change rates of each account status value. Specifically, accounts with larger absolute change rates can be ranked first, or the accounts with the fastest rate of decline can be assigned the highest priority. The purpose is to prioritize the accounts with the fastest deterioration. Determining recovery priority based on the comparison of updated account status values ​​of multiple physiological resource accounts means that when the differences in account status values ​​are not significant, the recovery order is directly ranked according to the current account status value of each account. Specifically, accounts with lower current status values ​​can be ranked first. The purpose is to prioritize the recovery of the accounts with the worst current status when the change trend is not obvious.

[0051] This application's solution introduces a judgment on the differences in account status values ​​and selects different priority strategies based on the judgment results, thereby more accurately reflecting the body's true recovery needs. Specifically, when it is determined that among the updated account status values ​​of multiple physiological resource accounts, at least two account status values ​​meet a preset numerical difference threshold, it indicates that the status values ​​of these accounts differ significantly, and some accounts may be significantly worse than others. In this case, simply comparing the current values ​​may not be sufficient to reveal the severity of the problem; for example, although one account's current value is not the lowest, its rate of decline is much faster than other accounts. Therefore, in this situation, the solution further obtains the historical account status values ​​of at least two physiological resource accounts that meet the numerical difference threshold within a preset time period and determines the rate of change of each account's status value based on these historical values. By comparing these rates of change, it is possible to identify which accounts are rapidly deteriorating, and thus, based on the comparison results of the rate of change of status values, determine the recovery priority, prioritizing the recovery of those accounts whose status is declining the fastest. This can more effectively prevent the further development of health risks.

[0052] In some preferred embodiments, a specific example is given below. Assume there are three physiological resource accounts: an energy account, a hydration account, and a core temperature account. Their account status values ​​represent the body's energy reserve level, hydration balance, and core temperature stability, respectively. The system obtains the updated account status values ​​for these three accounts, for example: energy account status value 60, hydration account status value 35, and core temperature account status value 85. A preset numerical difference threshold is set so that the difference between any two account status values ​​is greater than 20. First, it determines whether at least two of the three account status values ​​satisfy the numerical difference threshold condition. The difference between hydration account status value 35 and core temperature account status value 85 is 50, which is greater than 20, thus satisfying the condition. The difference between hydration account status value 35 and energy account status value 60 is 25, which is greater than 20, thus satisfying the condition. The difference between energy account status value 60 and core temperature account status value 85 is 25, which is greater than 20, thus satisfying the condition. Since there are account status value pairs that satisfy the numerical difference threshold condition (e.g., hydration account and core temperature account), the system proceeds to the historical data analysis phase. The system retrieves historical account status values ​​for the moisture and core temperature accounts within a preset time period (e.g., the past hour). Based on these historical values, it determines the rate of change of the moisture account's status value, for example, a decrease of 0.5 units per minute; and the rate of change of the core temperature account's status value, for example, a decrease of 0.1 units per minute. Based on the comparison of the rate of change of status values, the rate of decrease of the moisture account's status value (0.5) is faster than that of the core temperature account (0.1), therefore, the recovery priority of the moisture account is determined to be higher than that of the core temperature account. Although the energy account also meets the threshold condition, assuming its rate of change is a decrease of 0.3 units per minute, the recovery priority ranking may be: moisture account > energy account > core temperature account. Based on this, the recovery priority is determined, and a corresponding evaluation output is generated, such as prompting the user to replenish moisture first. As another specific implementation, assuming the updated account status values ​​of the three accounts are: energy account status value 70, moisture account status value 65, and core temperature account status value 72. The preset numerical difference threshold condition remains 20. It is determined whether at least two of the three account status values ​​meet the numerical difference threshold condition. The difference between the energy account and the water account is 5, not greater than 20; the difference between the energy account and the core temperature account is 2, not greater than 20; the difference between the water account and the core temperature account is 7, not greater than 20. Since there are no account state value pairs that meet the numerical difference threshold condition, the numerical difference threshold condition is not met. At this time, based on the comparison results of the updated account state values ​​of multiple physiological resource accounts, that is, by directly comparing the current values, the recovery priority is determined. The current state values ​​are sorted from low to high as follows: water account (65) < energy account (70) < core temperature account (72). Therefore, the recovery priority is determined as: water account > energy account > core temperature account.Based on this, recovery priorities are determined, and corresponding assessment outputs are generated, such as prompting users to replenish water first, followed by replenishing energy.

[0053] The above technical solution allows for flexible selection of strategies based on either current value comparison or historical rate of change comparison when determining the recovery priority of physiological resource accounts, depending on the magnitude of differences in account status values. When account status value differences are significant, analyzing historical trends can identify accounts deteriorating faster, allowing for priority recovery and preventing the neglect of potential risks by focusing solely on current values. When account status value differences are not significant, a simpler current value comparison strategy is employed. Therefore, this solution can more accurately determine the body's true recovery needs, providing more targeted and effective recovery guidance and avoiding unnecessary or inaccurate warnings and suggestions.

[0054] In some embodiments described above, a method is proposed to acquire external data corresponding to a specific load category within a load category, and to quantify the specific load category based on the external data to obtain a load quantification value. Specifically, this acquisition of external data corresponding to a specific load category and quantification based on the external data to obtain a load quantification value can be achieved by collecting external environmental information such as ambient temperature, humidity, and wind speed through sensors, or by collecting activity information such as activity type, intensity, and duration. Then, according to a preset quantification model or lookup table method, this external data is converted into a preliminary load quantification value. For example, for running activities, a running load value can be calculated based on data such as pace, incline, and duration; for high-temperature environments, a heat load value can be calculated based on temperature and humidity. This allows for a preliminary assessment of the load based on different external factors. However, in its implementation, simply using external data for quantification without considering the user's own physiological state may lead to a discrepancy between the quantification result and the user's actual load, thereby affecting the accuracy of health risk warnings.

[0055] To this end, this application further proposes the following steps for obtaining external data corresponding to a specific load category and quantifying the specific load category based on the external data to obtain a load quantization value:

[0056] Acquire external data corresponding to specific load categories and obtain physiological indicators reflecting user consumption status;

[0057] Based on external data, an initial load quantization value is generated;

[0058] Determine whether the initial load quantification value and physiological indicators meet the second preset threshold condition;

[0059] If the second preset threshold condition is met, a heat dissipation index that characterizes the body's heat dissipation state to the environment is determined based on physiological indicators and external environmental information in external data, and a corrected load quantization value is determined based on the heat dissipation index, so as to use the corrected load quantization value as the load quantization value.

[0060] If the second preset threshold condition is not met, the initial load quantization value will be used as the load quantization value.

[0061] The specific load category refers to a predefined class of activities or environmental factors that affect the user's physiological or psychological state. This can be defined using classification methods such as exercise load, environmental load, and psychological load. External data refers to information related to the specific load category and originating from outside the user's body. This can be obtained through environmental parameters collected by environmental sensors, user-inputted activity information, or task descriptions. Physiological indicators refer to biological measurements reflecting the user's internal physical state or response to external stimuli. These can be obtained using data collected by sensors such as heart rate, body surface temperature, and perspiration. The initial load quantification value is a preliminary numerical assessment of the specific load category based on external data. It can be generated using a preset formula or lookup table based on the external data. The second preset threshold condition is a preset judgment rule used to determine whether there is a significant difference between the initial load quantification value and the physiological indicators. This can be determined by comparing the difference, ratio, or trend of the two to see if it exceeds a preset range. External environmental information refers to the portion of external data related to the user's environment. This can be characterized using environmental parameters such as ambient temperature, humidity, wind speed, and solar radiation intensity. Among them, the heat dissipation index refers to a numerical assessment that characterizes the state or efficiency of the user's body in dissipating heat to the environment. It can be determined by calculating the heat loss rate or heat dissipation efficiency coefficient based on physiological and environmental data such as body surface temperature, ambient temperature, and perspiration. The corrected load quantification value refers to the value obtained by adjusting or recalculating the initial load quantification value after considering the user's physiological indicators and external environmental information. It can be determined by weighting or shifting the initial load quantification value based on the heat dissipation index, or by recalculating it through a new model.

[0062] This solution aims to more accurately quantify the impact of specific load categories on users. By comprehensively considering external data and user physiological indicators, the initial load quantification value is corrected, thereby improving the accuracy of the load quantification value. This solution lays the foundation for subsequent load quantification by acquiring external data corresponding to specific load categories and physiological indicators reflecting the user's consumption status. An initial load quantification value is generated based on the external data, which is a preliminary assessment of the load. The key is to determine whether the initial load quantification value and physiological indicators meet a second preset threshold condition. This judgment mechanism allows the system to identify situations where external data alone may not accurately reflect the user's actual physiological load. If the condition is met, it indicates a significant difference between the initial assessment and the user's actual physiological state. At this point, the solution further determines a heat dissipation indicator based on physiological indicators and external environmental information. The heat dissipation indicator characterizes the body's heat exchange state in the current environment and is an important manifestation of the impact of environmental factors on physiological load. Based on this heat dissipation indicator, a corrected load quantification value is determined, and the corrected value is used as the final load quantification value, thus making the quantification result closer to the user's true physiological burden in a specific environment. If the condition is not met, the initial load quantification value is considered to already reflect the user's actual situation well, and no correction is needed; it is directly used as the load quantification value. This is achieved through a dynamic correction mechanism based on physiological feedback and environmental factors.

[0063] In some preferred embodiments, this application is implemented as follows, taking an outdoor mountain bike riding scenario as an example, a specific load category can be defined as "environmental heat load". External data corresponding to the environmental heat load is acquired, such as current ambient temperature, relative humidity, and solar radiation intensity collected by sensors. Simultaneously, physiological indicators reflecting the user's consumption status are acquired, such as the user's body surface temperature, heart rate, and perspiration volume collected by sensors in sports underwear. Based on the acquired external data such as ambient temperature, relative humidity, and solar radiation intensity, an initial load quantification value can be generated, for example, calculating the current wet-bulb temperature (WBGT) or thermal index. Next, it is determined whether the initial load quantification value (WBGT) and the user's physiological indicators (such as body surface temperature and heart rate) meet a second preset threshold condition. For example, it can be determined whether there is a significant deviation between the current WBGT value and the increase in the user's body surface temperature or heart rate; if this deviation exceeds a preset range, the condition is considered met. If the second preset threshold condition is met, it indicates that the initial heat load value calculated solely from environmental data may not match the user's actual physiological response, and correction is required. Based on physiological indicators such as the user's body surface temperature, heart rate, and perspiration, as well as external environmental information such as ambient temperature, humidity, and wind speed, a heat dissipation index can be determined to characterize the body's heat dissipation to the environment, such as calculating the body's heat loss rate. Then, a corrected load quantization value is determined based on this heat dissipation index. For example, the initial WBGT value is adjusted according to the heat dissipation efficiency; a lower value is adjusted for high efficiency, and a higher value is adjusted for low efficiency. This corrected load quantization value is used as the final environmental heat load quantization value. If a second preset threshold condition is not met, the initial load quantization value is considered to reflect the user's actual situation well, and this initial load quantization value is directly used as the final environmental heat load quantization value.

[0064] The above technical solution considers not only external data but also physiological indicators reflecting the user's own consumption status when quantifying specific load categories. By judging the degree of matching between the initial load quantification value and the physiological indicators, potential deviations in quantification based solely on external data can be identified. When deviations exist, heat dissipation indicators are determined by combining physiological indicators and external environmental information, and the load quantification value is corrected based on these indicators. This allows the final load quantification value to more accurately reflect the user's actual physiological burden under specific environments and loads, overcoming the limitations of a single data source. Therefore, this solution provides more accurate load quantification results, offering a more reliable foundation for subsequent physiological resource consumption calculations and health risk warnings, thus improving the accuracy of health risk warnings.

[0065] In some embodiments of this application, a method is proposed to acquire external data corresponding to a specific load category and quantify the specific load category based on the external data to obtain a load quantification value. Specifically, this acquisition of external data corresponding to a specific load category and quantification based on the external data to obtain a load quantification value can be achieved by acquiring data such as speed and slope during exercise, and combining them with physiological indicators such as heart rate and body surface temperature to initially calculate an exercise load value. Then, the exercise load value is corrected based on the body's heat dissipation indicators to obtain a more accurate load quantification value. This allows for a comprehensive assessment of exercise load based on exercise intensity and the body's physiological response. However, in its implementation, the quantification value of a specific load category is only corrected based on heat dissipation indicators without fully considering other concurrent load factors, such as ambient temperature and humidity, which may result in an incomplete corrected load quantification value that cannot accurately reflect the overall pressure borne by the body, thereby affecting the reliability of subsequent health risk warnings.

[0066] In response, this application further proposes a step for determining a corrected load quantization value based on heat dissipation indicators, including:

[0067] Obtain the concurrent load quantization value for another load category that is different from the specific load category;

[0068] Based on the concurrent load quantization value, the preset mapping relationship between the heat dissipation index and the corrected load quantization value is adjusted to obtain an adjusted mapping relationship.

[0069] Based on the adjusted mapping relationship and heat dissipation index, the corrected load quantization value is determined.

[0070] Among them, the concurrent load quantification value refers to a numerical value obtained by quantifying other load factors that exist simultaneously with a specific load category and have an impact on the body. It can be calculated using a preset model or rules based on environmental sensor data (such as temperature, humidity, air pressure, wind speed, and solar radiation intensity) or user input information (such as the type of environment and altitude). The preset mapping relationship refers to the rules or functions that are established in advance to describe the correspondence between heat dissipation indicators and the corrected load quantification value without considering concurrent load. It can be stored in the form of lookup tables, mathematical formulas, or machine learning models. The adjusted mapping relationship refers to the new correspondence obtained after modifying the preset mapping relationship according to the concurrent load quantification value. It can be achieved by modifying the values ​​in the lookup table, adjusting the parameters in the mathematical formula, or updating the weights of the machine learning model.

[0071] This application's solution identifies and quantifies other load factors affecting the body by obtaining the concurrent load quantification value of a load category different from a specific load category. Based on this concurrent load quantification value, a preset mapping relationship between heat dissipation indicators and corrected load quantification values ​​is dynamically adjusted to obtain an adjusted mapping relationship. The preset mapping relationship describes the correspondence between heat dissipation indicators and corrected load quantification values ​​under a specific load category, but this preset relationship may no longer be accurate when concurrent loads are present. By adjusting the preset mapping relationship according to the concurrent load quantification value—for example, in high ambient temperatures, the body's heat dissipation capacity decreases, and even if the heat dissipation indicator shows good heat dissipation, the actual load quantification value should increase accordingly—the adjusted mapping relationship can more accurately reflect the actual significance of the heat dissipation indicator under the current comprehensive load situation. Finally, based on the adjusted mapping relationship and the heat dissipation indicator, the corrected load quantification value is determined. Since the mapping relationship has already considered the impact of concurrent loads, the obtained corrected load quantification value can more accurately reflect the comprehensive load borne by the body. The revised load quantification value is used as the load quantification value for subsequent consumption calculations of the account status value of the physiological resource account corresponding to a specific load category. This makes the status update of the physiological resource account more accurate. Finally, based on the more accurate account status value comparison results, an assessment output indicating the recovery priority is generated, which improves the reliability of health risk warning.

[0072] In some preferred embodiments, specifically, it is assumed that one load category is exercise load and another load category is ambient temperature load. First, the concurrent load quantization value of the ambient temperature load is obtained, for example, by acquiring the current ambient temperature through an environmental sensor and converting it into a quantified temperature load value. Then, based on this quantified temperature load value, a preset mapping relationship describing the relationship between exercise heat dissipation indicators (e.g., the difference between skin temperature and ambient temperature, perspiration rate) and the corrected exercise load quantization value is adjusted. For example, if the preset mapping relationship is a function `corrected exercise load = f(exercise heat dissipation indicator)`, when the temperature load quantization value is high, the parameters of function `f` can be adjusted to output a higher corrected exercise load value under the same exercise heat dissipation indicator, or a correction term related to the temperature load can be added to obtain the adjusted mapping relationship `corrected exercise load = g(exercise heat dissipation indicator, temperature load quantization value)`. Finally, the current exercise heat dissipation indicator and the temperature load quantization value are substituted into the adjusted mapping relationship `g` to calculate the final corrected exercise load quantization value.

[0073] The above technical solution can comprehensively consider the impact of other load categories that coexist with a specific load category, and dynamically adjust the mapping relationship between the heat dissipation index and the corrected load quantification value based on the concurrent load quantification value. This allows the corrected load quantification value to more accurately reflect the comprehensive load borne by the body, improves the accuracy of load quantification, and provides a more reliable basis for subsequent health risk warnings.

[0074] In some embodiments described above, a preset mapping relationship between heat dissipation indicators and corrected load quantization values ​​is adjusted based on concurrent load quantization values ​​to obtain an adjusted mapping relationship. Specifically, this adjustment can be achieved by scaling or shifting the preset mapping relationship based on the magnitude of the concurrent load quantization values ​​and consulting a preset adjustment coefficient table. This allows for preliminary correction of the relationship between heat dissipation indicators and corrected load quantization values ​​under different concurrent load conditions. However, this simple overall scaling or shifting approach may fail to capture more nuanced changes in the mapping relationship and accurately reflect the specific impact of different concurrent loads on the complex relationship between heat dissipation indicators and load quantization values, resulting in an inaccurate adjusted mapping relationship. Therefore, effectively adjusting the preset mapping relationship to ensure that the adjusted mapping relationship accurately reflects the relationship between heat dissipation indicators and corrected load quantization values ​​is a problem that needs to be solved.

[0075] In response, this application further proposes a step of adjusting the preset mapping relationship between the heat dissipation index and the corrected load quantization value based on the concurrent load quantization value to obtain an adjusted mapping relationship, including:

[0076] The preset mapping relationship is represented as a computational relationship determined by a set of functional parameters;

[0077] Based on the concurrent load quantization value, determine the adjustment value for at least one functional parameter in a set of functional parameters;

[0078] At least one functional parameter is updated based on the adjustment value, so that the calculation relationship determined by the updated set of functional parameters is used as the adjusted mapping relationship.

[0079] Among them, the preset mapping relationship refers to the preset association between the heat dissipation index, which characterizes the body's heat dissipation state to the environment, and the corrected load quantification value, which can be represented by functions, curves, lookup tables, or models; the functional parameters refer to the set of values ​​used to define the calculation relationship, which can be implemented by polynomial coefficients, the base of an exponential function, the weights of a neural network, or other adjustable parameters in the model; the calculation relationship refers to the mathematical or logical relationship between the corrected load quantification value and the heat dissipation index calculated through the functional parameters, which can be implemented by function formulas, algorithm models, or rule sets; the representation refers to transforming a form of mapping relationship into a form jointly described by the functional parameters and the calculation relationship; the concurrent load quantification value refers to the quantification value of another load category that is different from a specific load category, such as environmental load or psychological load; the adjustment value refers to the numerical increment or multiplier used to modify the functional parameters; determining the adjustment value refers to calculating or finding the specific value used to adjust the functional parameters based on the concurrent load quantification value; updating the functional parameters refers to modifying the existing value of the functional parameters based on the determined adjustment value; the adjusted mapping relationship refers to the new calculation relationship obtained after updating the functional parameters, which reflects the association between the adjusted heat dissipation index and the corrected load quantification value.

[0080] This application's solution represents a pre-defined mapping relationship as a computational relationship determined by a set of functional parameters, thus transforming what might otherwise be a complex mapping relationship into a set of quantifiable and adjustable parameters. This parameterized representation provides the foundation for subsequent dynamic adjustments. Different functional parameters correspond to different aspects of the mapping relationship, such as slope, intercept, and curvature. Based on this, and using the quantified concurrent load value, an adjustment value is determined for at least one functional parameter in the set. The quantified concurrent load value reflects the impact of other load categories, different from a specific load category, on the body. By considering the impact of these concurrent loads, it is possible to more accurately determine which aspects of the pre-defined mapping relationship need adjustment and to what extent. For example, if the concurrent load is high, it may be necessary to adjust the slope or intercept of the mapping relationship to change the degree of influence of the heat dissipation index on the corrected load quantification value. The method for determining the adjustment value can be selected according to the specific application scenario and requirements.

[0081] In some preferred embodiments, for example, the preset mapping relationship can be characterized as a simple linear function: Modified load quantization value = a * heat dissipation index + b. In this case, the functional parameters are the set {a, b}. Assume the concurrent load quantization value is the environmental load quantization value, which comprehensively reflects the additional effects of environmental temperature, humidity, wind speed, and other factors on the body. Based on the environmental load quantization value, adjustment values ​​for parameters a and / or b can be determined. For example, a rule can be preset: if the environmental load quantization value is high, a positive adjustment value is determined to increase parameter a (e.g., adjustment value _a = k * environmental load quantization value, where k is a preset coefficient), so that the influence weight of the heat dissipation index (e.g., the temperature difference between skin temperature and ambient temperature) on the modified load quantization value increases, reflecting the increased feeling of load due to the difficulty of heat dissipation in hot environments; simultaneously, a positive adjustment value can be determined to increase parameter b (e.g., adjustment value _b = m * environmental load quantization value, where m is a preset coefficient), to increase a basic correction amount, reflecting the additional load brought by the environment itself. Then, parameters a and b are updated based on the determined adjustment values, resulting in new parameters a' = a + adjustment value _a, b' = b + adjustment value _b. The calculation relationship determined by the updated parameter set {a', b'} (corrected load quantization value = a' * heat dissipation index + b') is used as the adjusted mapping relationship. In this way, the preset linear mapping relationship can be dynamically adjusted according to the actual environmental load, making it more adaptable to the current environmental conditions.

[0082] The above technical solution represents the preset mapping relationship as a computational relationship determined by a set of functional parameters. Based on the quantified concurrent load value, adjustments to the functional parameters are determined, and the functional parameters are then updated to obtain the adjusted mapping relationship. This method achieves refined and adaptive adjustment of the preset mapping relationship, more accurately reflecting the complex relationship between the body's heat dissipation state and actual load consumption under different concurrent load conditions. This helps to more accurately calculate the corrected load quantification value, thereby improving the accuracy and reliability of subsequent health risk assessments and recovery strategy formulation.

[0083] In some embodiments described above in this application, a heat dissipation index is proposed to determine a heat dissipation state of the body to the environment based on physiological indicators and external environmental information in external data, in order to correct the load quantification value. Specifically, the heat dissipation index can be determined by analyzing the relationship between physiological indicators such as heart rate and body surface temperature and external environmental information such as temperature and humidity. For example, in a high temperature and high humidity environment, heart rate and body surface temperature usually increase. The system can make a preliminary judgment on the body's heat dissipation state based on these increases, thus making a preliminary assessment of the impact of the environment on the body. However, in its implementation, the heat dissipation index is determined solely by physiological indicators and external environmental information without considering the impact of load unrelated to environmental factors on physiological indicators. For example, exercise intensity itself can also lead to an increase in heart rate. If the physiological changes caused by exercise intensity are not distinguished, the heat dissipation index may be inaccurate, thereby affecting the correction effect of the load quantification value.

[0084] In this regard, this application further proposes a step for determining a heat dissipation index characterizing the body's heat dissipation state to the environment based on physiological indicators and external environmental information from external data, including:

[0085] Obtain state information representing loads independent of environmental factors;

[0086] Based on the state information, a baseline physiological impact attributable to the load unrelated to environmental factors is determined;

[0087] Based on the physiological indicators and the baseline physiological influence, the physiological response components caused by environmental factors in the physiological indicators are separated.

[0088] The heat dissipation index is determined based on the physiological response components and the external environmental information.

[0089] Among them, the state information characterizing the load unrelated to environmental factors refers to data reflecting the degree of physiological load generated by the body due to its own activities and not caused by environmental factors. This can be achieved using information such as exercise intensity (pace, power, heart rate zone, exercise type, exercise duration, body posture, metabolic rate, etc.), with the aim of quantifying the impact of internally generated load on physiological indicators. The baseline physiological impact attributable to the load unrelated to environmental factors refers to the expected change in physiological indicators caused by a specific load unrelated to environmental factors under ideal conditions excluding the influence of environmental factors. This can be calculated based on historical data, physiological models, or preset rules. For example, at a specific exercise intensity, the expected changes in heart rate, body temperature, or metabolic rate calculated by looking up tables or formulas. Its purpose is to establish a reference baseline for subsequent separation of physiological responses caused by environmental factors. The physiological response component refers to the component of physiological indicators attributable to environmental factors. For example, physiological changes caused directly or indirectly by factors such as temperature, humidity, wind speed, and altitude, after eliminating or weakening the influence of loads unrelated to environmental factors, can be calculated using mathematical models such as linear models, nonlinear models, statistical methods, or machine learning algorithms by comparing actual physiological indicators with baseline physiological impact quantities. The purpose is to accurately quantify the specific impact of environmental factors on the body's physiological state. Heat dissipation indicators refer to a comprehensive numerical or state description that characterizes the efficiency or state of the body in dissipating heat to the external environment under current environmental conditions. They can be determined by calculation models or lookup tables based on physiological response components such as changes in body surface temperature and sweat excretion rate, and external environmental information such as ambient temperature, humidity, and wind speed. For example, it can be a heat dissipation efficiency score, heat stress index, or heat dissipation capacity level. The purpose is to assess the body's ability to cope with environmental heat load and provide a basis for correcting the load quantification value.

[0090] This application's solution obtains state information characterizing loads unrelated to environmental factors. Based on this state information, a baseline physiological impact quantity attributable to the load unrelated to environmental factors is determined, thus establishing a reference baseline. This baseline reflects the expected impact of non-environmental loads on physiological indicators when there is no environmental interference. Next, based on the actual physiological indicators and the baseline physiological impact quantity, the physiological response component caused by environmental factors is separated from the physiological indicators. This step, by comparing the actual physiological indicators with the baseline expected value, effectively eliminates the interference of non-environmental factors on the physiological indicators, allowing the separated physiological response component to more accurately reflect the impact of environmental factors on the body's physiological state. Finally, a heat dissipation index is determined based on the separated physiological response component and external environmental information. Since the physiological response component has excluded the influence of loads unrelated to environmental factors, the heat dissipation index determined in conjunction with external environmental information can more accurately characterize the body's heat dissipation state to the environment. In the health risk early warning method, when the initial load quantification value and physiological indicators meet preset conditions, the heat dissipation index is determined based on the physiological indicators and external environmental information, and the initial load quantification value is corrected based on this heat dissipation index. The more accurate heat dissipation index determination method provided by this solution can obtain more precise heat dissipation indexes, which can then be used to more accurately correct load quantification values, thereby improving the accuracy of overall health risk warning.

[0091] In some preferred embodiments, such as in outdoor sports scenarios, the user's current exercise intensity, such as pace, can be acquired as state information representing the load unrelated to environmental factors. Based on this pace information, an expected heart rate value at that pace can be determined according to a preset exercise physiology model or the user's historical data, serving as a baseline physiological impact attributable to the load unrelated to environmental factors. Then, the user's actual monitored heart rate is compared with the expected heart rate value. A calculation model, such as a simple difference calculation or a more complex nonlinear model, is used to calculate the portion of the actual heart rate exceeding the expected value, or the model outputs a value representing the additional heart rate increase caused by environmental factors such as high temperature or high humidity, which is used as a physiological response component. Finally, combining this physiological response component, such as the environmentally induced heart rate increase, with current external environmental information such as ambient temperature and humidity, a heat dissipation index representing the body's heat dissipation status is determined using a preset algorithm or lookup table method. For example, if the environmentally induced heart rate increase is large and the ambient temperature is high, the heat dissipation index may indicate a heavy heat dissipation burden on the body.

[0092] By employing the aforementioned technical solution, when determining the heat dissipation index characterizing the body's heat dissipation from the environment, the influence of loads unrelated to environmental factors on physiological indicators is considered and separated from the actual physiological indicators. This allows for a more accurate acquisition of the physiological response components caused by environmental factors. The heat dissipation index determined based on more accurate physiological response components and external environmental information can more precisely characterize the body's heat dissipation from the environment, and can thus be used to more accurately correct load quantification values, improving the accuracy of health risk warnings.

[0093] In some embodiments described above in this application, a method is proposed to separate the physiological response component caused by environmental factors from physiological indicators based on physiological indicators and baseline physiological influence quantities, in order to more accurately assess the impact of environmental factors on the body. Specifically, this separation of physiological response components can be achieved by using a linear regression model, with physiological indicators as dependent variables and baseline physiological influence quantities as independent variables, and estimating the physiological response component caused by environmental factors through linear fitting. This can initially distinguish the influence of different factors on physiological indicators. However, in its implementation, the relationship between physiological indicators and baseline physiological influence quantities is often non-linear. Simply using a linear model may not accurately separate the physiological response component, leading to deviations in the subsequent calculation of heat dissipation indicators, and thus affecting the accuracy of health risk warnings.

[0094] In this regard, this application further proposes a step for separating the physiological response component caused by environmental factors from the physiological indicators based on physiological indicators and baseline physiological influence quantities, including:

[0095] A pre-defined nonlinear model is used;

[0096] Physiological indicators and baseline physiological influence quantities are used as inputs to the nonlinear model;

[0097] The physiological response components are determined based on the output of the nonlinear model.

[0098] Specifically, this separation step involves the following technical features: the preset nonlinear model refers to a mathematical model or computational structure that can describe the nonlinear relationship between input and output, which can be implemented using neural network models, support vector regression models, decision tree models, kernel function methods, etc.; the input of the nonlinear model refers to the data provided to the nonlinear model for processing, specifically physiological indicators and baseline physiological influence quantities; the output of the nonlinear model refers to the result calculated by the nonlinear model based on the input data; and the physiological response component refers to the physiological change component in the physiological indicators that is independently caused by environmental factors.

[0099] Based on the aforementioned technical features, the solution of this application employs a pre-defined nonlinear model, using physiological indicators and baseline physiological impact quantities as model inputs. The nonlinear model can capture the complex nonlinear relationship between physiological indicators and baseline physiological impact quantities, performing nonlinear mapping processing on the input data. The model's output directly characterizes the portion of the physiological indicator that is separated from the baseline physiological impact quantity (i.e., the load unrelated to environmental factors), thereby determining the physiological response component caused by environmental factors. This heat dissipation indicator is further used to correct the initial load quantification value, obtaining a more accurate load quantification value. The corrected load quantification value is used for calculating the consumption and updating the status of the physiological resource account, ultimately generating a health risk warning assessment output based on the account status value.

[0100] In some preferred embodiments, the separation step described above can be specifically implemented as follows: the preset nonlinear model can be a feedforward neural network. This neural network can include an input layer, at least one hidden layer, and an output layer. The input layer is configured to receive physiological indicators and baseline physiological influence quantities as input data. The hidden layer is configured to contain a nonlinear activation function for learning the nonlinear relationship between the input data. The output layer is configured to output one or more values ​​used to determine the physiological response components. For example, the neural network can directly output the increment or decrement in the physiological indicators caused by environmental factors, i.e., the physiological response components. This neural network can be trained using historically collected physiological indicators, baseline physiological influence quantities, and corresponding physiological response data under the influence of environmental factors to optimize the model's parameters, enabling it to accurately separate the physiological response components caused by environmental factors from the physiological indicators.

[0101] By employing the aforementioned technical solution and a pre-defined nonlinear model, the complex nonlinear relationship between physiological indicators and baseline physiological influence quantities can be captured more accurately, thereby more precisely separating the physiological response components caused by environmental factors. This provides a more accurate data foundation for subsequent heat dissipation index calculations, thereby improving the accuracy of load quantification correction and ultimately making the health risk warning assessment output more reliable.

[0102] In some of the embodiments described above in this application, when determining the heat dissipation index, it is proposed to determine the heat dissipation index based on physiological response components and external environmental information. Specifically, the heat dissipation index can be calculated solely based on the current physiological response components and external environmental information. This allows for a rapid response to changes in the current environment and physiological state. However, relying solely on the current physiological response components and external environmental information may not accurately reflect the individual differences and historical environmental adaptability of users, leading to a decrease in the accuracy of the heat dissipation index and consequently affecting the accuracy of health risk warnings.

[0103] In this regard, this application further proposes a step for determining heat dissipation indicators based on physiological response components and external environmental information, including:

[0104] Acquire the user's historical physiological response components and historical external environment information;

[0105] Based on the historical physiological response components and the historical external environment information, the relationship between the user-specific physiological response components, the external environment information, and the heat dissipation index is determined.

[0106] The heat dissipation index is determined based on the user-specific relationship, the current physiological response components, and the external environmental information.

[0107] The historical physiological response component refers to the set of physiological response data caused by environmental factors, separated from the user's physiological indicators and baseline physiological influence quantities over a past period. Specifically, it can include environmentally related physiological response data such as heart rate changes, body surface temperature changes, and sweat secretion rate under different historical environmental conditions. Its purpose is to provide a historical record of an individual's physiological adaptation to different environmental stimuli. The historical external environmental information refers to the set of external environmental conditions the user was in during the time period corresponding to the aforementioned historical physiological response component. Specifically, it can include historical environmental parameters such as temperature, humidity, wind speed, solar radiation intensity, and altitude. Its purpose is to establish a mathematical model or mapping rule that can characterize the individual user's response characteristics to environmental heat load. This can be achieved using statistical regression models, machine learning algorithms, or parameter fitting based on physiological models. Its purpose is to capture the differences in heat dissipation capacity and adaptability of individual users under different environmental conditions, thereby enabling personalized calculation of heat dissipation indicators.

[0108] The proposed solution acquires historical physiological response components and historical external environmental information of a specific user, and based on this historical data, establishes a specific relationship between the physiological response components, external environmental information, and heat dissipation indicators that reflects the individual characteristics of the user. This specific relationship captures the user's long-term physiological adaptation patterns and individual differences under different environmental conditions. Subsequently, during real-time monitoring, the current physiological response components and external environmental information are input into the user-specific relationship model to calculate a more accurate heat dissipation indicator.

[0109] In some preferred embodiments, the system continuously collects the user's physiological response components and corresponding external environmental information under different exercise and environmental conditions, and stores this historical data in the user's personal data profile. For example, it can record the user's heart rate increase relative to a baseline heart rate (physiological response component) when exercising in a high-temperature and high-humidity environment, as well as the ambient temperature and humidity (external environmental information). The system can periodically, or after accumulating a certain amount of historical data, use this data to train a personalized prediction model for the user, for example, using a linear regression model or a more complex machine learning model. The input features of this model include physiological response components and external environmental information, and the output is a predicted heat dissipation index. Through training with historical data, the model parameters are optimized to best fit the relationship between the user's past physiological responses and the environment. When the user engages in new exercise, the system acquires the current physiological response components and external environmental information in real time and inputs this real-time data into the user-specific prediction model. The model output is the heat dissipation index at the current moment.

[0110] The above technical solution fully considers the user's historical physiological response and environmental adaptability when determining heat dissipation indicators. By establishing the relationship between user-specific physiological response components, external environmental information, and heat dissipation indicators, it is possible to more accurately assess an individual's actual heat dissipation status in a specific environment. This overcomes the limitations of relying solely on current data to calculate heat dissipation indicators, improves the calculation accuracy of heat dissipation indicators, and thus provides a more reliable data foundation for subsequent health risk assessment and early warning, enhancing the pertinence and effectiveness of health risk early warning.

[0111] As a specific example, the multi-data fusion health risk warning of this application can be implemented according to the following steps:

[0112] The implementation of this method involves three core steps: load classification and quantification, associated consumption of recovery resources, and guidance generation based on account status.

[0113] Step 1: Classification and quantification of load;

[0114] The system pre-sets several main load types, including mechanical load, thermal load, and altitude load. During operation, the system periodically (e.g., every minute) categorizes and quantifies the currently generated load based on the acquired information.

[0115] 1. Quantification of Mechanical Load: This mainly relates to the physical work done by the athlete. The system acquires speed and elevation gain information through connected devices (such as bicycle speedometers or mobile phone GPS). The quantification rule can be simplified as follows: the mechanical load value equals A multiplied by (vertical elevation gain in meters) plus B multiplied by (equivalent distance on flat ground). Here, A and B are preset coefficients.

[0116] 2. Quantification of Heat Load: This mainly relates to the physiological cost the human body pays to combat high-temperature environments. The system acquires the external ambient temperature and the skin temperature collected by the underwear. The quantification rule can be: the heat load value equals C multiplied by (ambient temperature minus the comfort temperature threshold) multiplied by the duration. The comfort temperature threshold (e.g., 22℃) can be preset or user-defined, and C is a coefficient that can be appropriately increased when the skin temperature remains above a certain level.

[0117] 3. Quantification of Altitude Load: This mainly relates to the additional stress that high altitude places on the body. When the altitude exceeds a safe threshold (e.g., 1800 meters), the system begins to calculate this load: the altitude load value equals D multiplied by (current altitude minus altitude threshold) multiplied by the duration, where D is a coefficient.

[0118] Step 2: Restore associated resource consumption;

[0119] The system maintains a virtual "recovery resource ledger" containing several independent accounts, primarily including: an "energy reserve account," a "fluid balance account," and a "thermoregulation account." Each account has an initial value (e.g., 100) at the start of exercise.

[0120] The system correlates the different loads quantified in the first step with the consumption of these recovery resources:

[0121] 1. Mechanical load mainly consumes the "energy reserve account". The system will deduct the calculated mechanical load value from the "energy reserve account" according to a certain conversion relationship.

[0122] 2. The heat load primarily consumes the "fluid balance account" and the "thermoregulation account." The system deducts the calculated heat load value from these two accounts according to preset proportions. For example, 70% of the consumption is recorded under the "fluid balance account," and 30% under the "thermoregulation account."

[0123] 3. Altitude load, as a systemic pressure, accelerates the consumption of other resources. This consumption can occur by using the calculated altitude load value as a multiplier to slightly increase the deductions from the corresponding accounts for mechanical and thermal loads within the same time period.

[0124] Before deducting from the account, the system performs a simple reading rationality check. For example, when the sweat sensor reading shows a sudden change that is inconsistent with the trend of heart rate and body temperature changes, the system will temporarily ignore the sweat information for that period and instead estimate the consumption of the "fluid balance account" based on the heat load calculation results, thereby avoiding the impact of sensor instability.

[0125] Step 3: Generating instructions based on account status;

[0126] 1. Real-time Risk Warning: During exercise, the system continuously monitors the "balance" of all recovery resource accounts. When the balance of any account falls below a preset warning line (e.g., 30), the system will issue a targeted warning. For example, if the "fluid balance account" balance is too low, the system will prompt: "Increased risk of dehydration. The current environment is causing significant water loss to the body. Please replenish with electrolyte-containing drinks immediately."

[0127] 2. Post-exercise Recovery Guidance: After exercise, the system generates a priority recovery report based on the final balance of each recovery resource account. The report first lists the accounts most severely depleted and provides primary recovery recommendations. For example, if the final balance of the "Energy Reserve Account" is 15 and the "Fluid Balance Account" is 40, the report will state: "This exercise posed the greatest challenge to your energy reserves. Primary recovery task: Replenish with easily absorbed carbohydrates within 30 minutes after exercise. Secondary task: Continuously sip water to restore fluid balance."

[0128] This method categorizes these stress sources into several indirectly quantifiable types, such as "mechanical load" and "thermal load." For example, it quantifies the mechanical work done by climbing using GPS information and the thermal stress caused by high temperatures using ambient temperature information.

[0129] Subsequently, this method established a one-to-one correspondence, linking different types of workload expenditures to the consumption of specific "recovery resources" (such as energy and body fluids) in the body. This is like creating a ledger for the body, where every hill climb consumes the "energy reserve account," and every ride in high temperatures consumes the "fluid balance account."

[0130] By continuously tracking changes in the "balance" of these virtual accounts, the system bypasses the direct interpretation of single, vague physiological indicators (such as heart rate). Even with the same high heart rate, the system can determine whether the surge in expenditure from the "energy reserve account" (due to high-intensity exercise) or the "fluid balance account" (due to heat stress) is due to a review of the ledger. Based on this assessment, the system can issue highly targeted warnings and, after exercise, generate a prioritized recovery "repayment" plan based on the final "debt" amount in each account. This enables precise and effective health risk management and recovery guidance under complex conditions.

[0131] To illustrate this method more clearly, we will use a mountain bike enthusiast as an example.

[0132] Scenario: The user plans a 3-hour mountain bike ride. The first 1.5 hours are uphill climbs through forests at 15°C, and the last 1.5 hours are flat rides on a ridge with a temperature of 33°C and strong sunlight. He is wearing a sports bra with integrated heart rate and body temperature sensors, connected to a mobile app that provides GPS and weather information.

[0133] Situations where this method is used:

[0134] The user's app has three internal recovery resource accounts, each with an initial value of 100 points: energy reserve account, fluid balance account, and body temperature regulation account.

[0135] 1. The exercise begins, and all accounts start with 100 points.

[0136] 2. First half of the forest climb (0 to 1.5 hours):

[0137] The system detected significant altitude gain (high mechanical load) and low ambient temperature (no thermal load). Therefore, the system primarily deducted points from the "Energy Reserve Account." After 1.5 hours, the account status became: Energy Reserve Account: 40 points, Fluid Balance Account: 85 points, Thermoregulation Account: 95 points. At this point, energy consumption was the primary concern.

[0138] 3. Ridge cycling (second half) (1.5 to 3 hours):

[0139] The system detected minimal altitude gain (low mechanical load) but an ambient temperature as high as 33°C (high thermal load). Therefore, the system began deducting points primarily from the "Fluid Balance Account" and "Thermoregulation Account," while also deducting a small amount from the "Energy Reserve Account."

[0140] After 2.5 hours of cycling, the "fluid balance account" score dropped to 28 points, below the warning threshold of 30 points. The app immediately issued a highly specific voice prompt: "Warning: Extremely high risk of dehydration. The current high temperature environment is rapidly depleting your body fluids. Please drink electrolyte-containing water immediately; plain water is not effective." This prompt accurately pointed out the root cause of the problem.

[0141] 4. At the end of the exercise, the final status of each account is as follows: Energy Reserve Account: 25 points, Fluid Balance Account: 10 points, Thermoregulation Account: 60 points.

[0142] 5. The system-generated recovery report clearly indicates the priorities:

[0143] "Guidelines for recovery from this cycling trip:"

[0144] First priority (most urgent): Your fluid balance is severely depleted. Immediately replenish with at least 500 ml of electrolyte-containing sports drink to avoid the health risks associated with severe dehydration.

[0145] Secondary task: Your energy reserves are greatly depleted. Within 30 minutes of rehydrating, consume easily absorbed carbohydrates (such as energy gels or bananas) to initiate the recovery process.

[0146] General tasks: Pay attention to cooling down your body and avoid prolonged exposure to high temperatures.

[0147] As can be seen from this embodiment, this method successfully solves the problem that traditional methods cannot distinguish the nature of stress in complex environments by decomposing the sources of stress and independently accounting for different recovery resources. This provides a decisive, accurate and prioritized risk warning and recovery strategy.

[0148] Secondly, referring to Figure 2 This application further proposes a multi-data fusion health risk early warning system, which includes:

[0149] The storage module 201 is defined to predefine multiple load categories and multiple physiological resource accounts corresponding to the load categories;

[0150] The data acquisition and quantization module 202 is used to acquire external data corresponding to a specific load category in the load category, and to quantize the specific load category based on the external data to obtain a load quantization value;

[0151] The consumption calculation and update module 203 is used to calculate the consumption of the account status value of the physiological resource account corresponding to the specific load category based on the load quantification value, so as to update the account status value.

[0152] The evaluation output module 204 is used to generate an evaluation output indicating the recovery priority based on the comparison results of the updated account status values ​​of the plurality of physiological resource accounts when the preset evaluation conditions are met.

[0153] The above technical solution provides a multi-data fusion health risk early warning system, which serves as a specific implementation carrier for the multi-data fusion health risk early warning method, enabling the method to be effectively executed and applied.

[0154] The above description is merely an embodiment of this application and is not intended to limit the scope of protection of this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of protection of this application.

Claims

1. A multi-data fusion health risk early warning method, characterized in that, include: Multiple load categories and multiple physiological resource accounts corresponding to the load categories are predefined; Obtain external data corresponding to a specific load category in the load categories, and quantize the specific load category based on the external data to obtain a load quantization value; Based on the load quantification value, the account status value of the physiological resource account corresponding to the specific load category is calculated to update the account status value. When the preset evaluation conditions are met, an evaluation output indicating the recovery priority is generated based on the comparison results of the updated account status values ​​of the multiple physiological resource accounts. The step of calculating the consumption of the account status value of the physiological resource account corresponding to the specific load category based on the load quantification value, and updating the account status value, includes: Obtain physiological indicators that reflect the user's consumption status; Determine whether the load quantification value and the physiological indicator meet the first preset threshold condition; If the first preset threshold condition is met, an adjusted resource consumption value is determined based on the physiological indicators, and the account status value of the physiological resource account is calculated based on the adjusted resource consumption value. If the first preset threshold condition is not met, then the consumption is calculated for the account status value of the physiological resource account corresponding to the specific load category based on the load quantification value. The step of generating an assessment output indicating recovery priority based on a comparison of the updated account status values ​​of the plurality of physiological resource accounts includes: Determine whether there are at least two account status values ​​among the updated account status values ​​of the multiple physiological resource accounts that satisfy a preset numerical difference threshold condition; If the numerical difference threshold condition is met, the historical account status values ​​of at least two physiological resource accounts that meet the numerical difference threshold condition are obtained within a preset time period. Based on the historical account status values, the rate of change of the status value of each account is determined. Based on the comparison results of the rate of change of the status value, the recovery priority is determined, and the evaluation output is generated. If the numerical difference threshold condition is not met, the recovery priority is determined based on the comparison results of the updated account status values ​​of the multiple physiological resource accounts, and the evaluation output is generated. The step of acquiring external data corresponding to a specific load category in the load category, and quantizing the specific load category based on the external data to obtain a load quantization value, includes: Obtain the external data corresponding to the specific load category, and obtain physiological indicators reflecting the user's consumption status; Based on the external data, an initial load quantization value is generated; Determine whether the initial load quantification value and the physiological indicators meet the second preset threshold condition; If the second preset threshold condition is met, a heat dissipation index representing the body's heat dissipation state to the environment is determined based on the physiological index and the external environmental information in the external data, and a corrected load quantization value is determined based on the heat dissipation index, so as to use the corrected load quantization value as the load quantization value. If the second preset threshold condition is not met, the initial load quantization value will be used as the load quantization value.

2. The multi-data fusion health risk early warning method according to claim 1, characterized in that, The step of determining a corrected load quantization value based on the heat dissipation index includes: Obtain the concurrent load quantization value of another load category that is different from the specific load category; Based on the concurrent load quantization value, the preset mapping relationship between the heat dissipation index and the corrected load quantization value is adjusted to obtain an adjusted mapping relationship; Based on the adjusted mapping relationship and the heat dissipation index, the corrected load quantization value is determined.

3. The multi-data fusion health risk early warning method according to claim 2, characterized in that, The step of adjusting the preset mapping relationship between the heat dissipation index and the corrected load quantization value based on the concurrent load quantization value to obtain an adjusted mapping relationship includes: The preset mapping relationship is represented as a computational relationship determined by a set of functional parameters; Based on the concurrent load quantization value, determine the adjustment value for at least one functional parameter in the set of functional parameters; The at least one functional parameter is updated according to the adjustment value, so that the calculation relationship determined by the updated set of functional parameters is used as the adjusted mapping relationship.

4. The multi-data fusion health risk early warning method according to claim 1, characterized in that, The step of determining a heat dissipation index characterizing the body's heat dissipation state to the environment based on the physiological indicators and the external environmental information in the external data includes: Obtain state information representing loads independent of environmental factors; Based on the state information, a baseline physiological impact attributable to the load unrelated to environmental factors is determined; Based on the physiological indicators and the baseline physiological influence, the physiological response components caused by environmental factors in the physiological indicators are separated. The heat dissipation index is determined based on the physiological response components and the external environmental information.

5. The multi-data fusion health risk early warning method according to claim 4, characterized in that, The step of separating the physiological response component caused by environmental factors from the physiological indicators based on the physiological indicators and the baseline physiological influence quantity includes: A pre-defined nonlinear model is used; The physiological indicators and the baseline physiological influence quantities are used as inputs to the nonlinear model; The physiological response components are determined based on the output of the nonlinear model.

6. The multi-data fusion health risk early warning method according to claim 4, characterized in that, The step of determining the heat dissipation index based on the physiological response components and the external environmental information includes: Acquire the user's historical physiological response components and historical external environment information; Based on the historical physiological response components and the historical external environment information, the relationship between the user-specific physiological response components, the external environment information, and the heat dissipation index is determined. The heat dissipation index is determined based on the user-specific relationship, the current physiological response components, and the external environmental information.

7. A multi-data fusion health risk early warning system, characterized in that, The system includes: Define a storage module for pre-defining multiple load categories and multiple physiological resource accounts corresponding to the load categories; The data acquisition and quantization module is used to acquire external data corresponding to a specific load category in the load category, and to quantize the specific load category based on the external data to obtain a load quantization value; The consumption calculation and update module is used to calculate the consumption of the account status value of the physiological resource account corresponding to the specific load category based on the load quantification value, so as to update the account status value. The evaluation output module is used to generate an evaluation output indicating the recovery priority based on the comparison results of the updated account status values ​​of the multiple physiological resource accounts when the preset evaluation conditions are met. The step of calculating the consumption of the account status value of the physiological resource account corresponding to the specific load category based on the load quantification value, and updating the account status value, includes: Obtain physiological indicators that reflect the user's consumption status; Determine whether the load quantification value and the physiological indicator meet the first preset threshold condition; If the first preset threshold condition is met, an adjusted resource consumption value is determined based on the physiological indicators, and the account status value of the physiological resource account is calculated based on the adjusted resource consumption value. If the first preset threshold condition is not met, then the consumption is calculated for the account status value of the physiological resource account corresponding to the specific load category based on the load quantification value. The step of generating an assessment output indicating recovery priority based on a comparison of the updated account status values ​​of the multiple physiological resource accounts includes: Determine whether there are at least two account status values ​​among the updated account status values ​​of the multiple physiological resource accounts that satisfy a preset numerical difference threshold condition; If the numerical difference threshold condition is met, the historical account status values ​​of at least two physiological resource accounts that meet the numerical difference threshold condition are obtained within a preset time period. Based on the historical account status values, the rate of change of the status value of each account is determined. Based on the comparison results of the rate of change of the status value, the recovery priority is determined, and the evaluation output is generated. If the numerical difference threshold condition is not met, the recovery priority is determined based on the comparison results of the updated account status values ​​of the multiple physiological resource accounts, and the evaluation output is generated. The step of acquiring external data corresponding to a specific load category in the load category, and quantizing the specific load category based on the external data to obtain a load quantization value includes: Obtain the external data corresponding to the specific load category, and obtain physiological indicators reflecting the user's consumption status; Based on the external data, an initial load quantization value is generated; Determine whether the initial load quantification value and the physiological indicators meet the second preset threshold condition; If the second preset threshold condition is met, a heat dissipation index representing the body's heat dissipation state to the environment is determined based on the physiological index and the external environmental information in the external data, and a corrected load quantization value is determined based on the heat dissipation index, so as to use the corrected load quantization value as the load quantization value. If the second preset threshold condition is not met, the initial load quantization value will be used as the load quantization value.

Images