Construction method of stroke prevention map system based on artificial intelligence

Abstract

The invention relates to the technical field of construction of prevention map systems, in particular to a construction method of a stroke prevention map system based on artificial intelligence. The method comprises the following steps that real-time physiological sign data of a stroke patient is collected through a wearable device, then historical case data is collected, the environment where the patient is located is extracted, case trigger inducement analysis is conducted, and stroke patient trigger inducement data is obtained; performing stroke sudden risk probability calculation on the stroke patient triggering inducement data, and performing stroke local risk environment level calculation to obtain a stroke local risk environment level; constructing a dynamic stroke prevention and early warning map according to the stroke local risk environment level to obtain the dynamic stroke prevention and early warning map; and carrying out risk assessment result response on the dynamic stroke prevention early warning map to obtain risk assessment result response data. According to the method, the construction technology of the preventive map system is optimized, so that the construction technology of the preventive map system is more perfect.

Description

Technical Field

[0001] The present invention relates to the technical field of prevention map system construction, and in particular to a method for constructing a stroke prevention map system based on artificial intelligence. Background Art

[0002] As one of the leading causes of death and disability worldwide, stroke has become a major problem that needs to be solved urgently in the field of public health. The incidence, disability and mortality of stroke have shown an increasing trend year by year, especially in developing countries, where the incidence of stroke remains high. Studies have shown that the occurrence of stroke is closely related to multiple factors, including genetics, lifestyle, environment, diet, blood pressure management and other variable factors. Although modern medicine has made certain progress in the prevention, diagnosis and treatment of stroke, the early symptoms of stroke are not obvious and occur rapidly, and many people lack timely prevention and intervention measures, resulting in high morbidity and mortality. Therefore, how to reduce the incidence of stroke through effective preventive measures, early screening and real-time monitoring has become a key issue in medical research and public health management. Against this background, a stroke prevention map system based on artificial intelligence (AI) technology came into being. The system is committed to providing accurate stroke risk prediction and personalized prevention recommendations by combining big data analysis, geographic information system (GIS) and artificial intelligence algorithms. AI technology can perform efficient disease risk assessment and pattern recognition by analyzing multidimensional data such as individual health records, living habits, genetic information, and environmental factors. In addition, combined with GIS technology, the stroke prevention map system can effectively associate various disease data with geographic locations to generate a stroke risk distribution map with dynamic spatial and temporal changes. However, the traditional method of constructing a stroke prevention map system based on artificial intelligence has the problems of low accuracy in stroke risk prediction and recognition, and the inability to accurately warn of risk trends. Summary of the invention

[0003] Based on this, it is necessary to provide a method for constructing an artificial intelligence-based stroke prevention map system to solve at least one of the above technical problems.

[0004] To achieve the above purpose, a method for constructing a stroke prevention map system based on artificial intelligence is provided, and the method comprises the following steps: Step S1: collecting real-time physiological sign data of stroke patients through wearable devices to obtain real-time sign filling data of stroke patients; collecting historical case data of stroke patients to obtain desensitized historical case data of stroke patients; Step S2: extracting the patient's environment from the real-time vital sign filling data of the stroke patient to obtain the patient's environment data; analyzing the case triggering factors of the stroke patient's historical case desensitization data to obtain the stroke patient's triggering factor data; calculating the stroke sudden risk probability of the stroke patient's triggering factor data to obtain the stroke sudden risk probability data; calculating the stroke local risk environment level of the patient's environment data based on the stroke sudden risk probability data to obtain the stroke local risk environment level; Step S3: performing a geographical environment stroke triggering trend analysis on the stroke sudden risk probability data according to the stroke local risk environment level to obtain geographical environment stroke triggering trend data; constructing a dynamic stroke prevention and early warning map based on the geographical environment stroke triggering trend data to obtain a dynamic stroke prevention and early warning map; Step S4: respond to the risk assessment results of the dynamic stroke prevention and early warning map based on the stroke triggering trend data of the geographical environment to obtain risk assessment result response data, and use the preset medical resource matching model to push the risk assessment result response data to the referral agency to obtain stroke risk referral agency push data.

[0005] The present invention collects real-time physiological sign data of stroke patients through wearable devices, and can obtain important physiological parameters such as the patient's current health status, heart rate, blood pressure, body temperature, etc., to ensure that the patient's physical condition is grasped in time. In addition, by collecting desensitized data of patients' historical cases, it can provide a comprehensive understanding of the patient's previous health status, including medical history, medication history, previous stroke attacks and other information. This lays the foundation for subsequent risk assessment and the formulation of personalized treatment plans. By extracting the environment of the patient's real-time physiological data, it is possible to accurately assess the risk factors of the patient's environment, such as climate change, air quality, etc., which have a direct impact on the risk of stroke. Secondly, based on historical case data, the triggering factors of stroke patients (such as diet, mood, physical activity, unstable blood pressure, increased blood sugar, hypertension, etc.) are analyzed, and then the risk probability of sudden stroke is calculated. This analysis provides quantitative data for determining the patient's potential stroke risk, and evaluates the safety of the patient's current environment by calculating the local risk environment level. Based on the stroke sudden risk probability data, the relationship between the geographical environmental factors and the stroke triggering trend in the patient's area is analyzed to obtain the geographical environment stroke triggering trend data. This trend data helps identify the incidence pattern of stroke in a specific geographic area, such as certain geographical locations that have an increased risk of stroke due to environmental pollution or climate change. Based on these analysis results, a dynamic stroke prevention and warning map is constructed to enable medical institutions, patients and their families to understand the stroke risk status of each region in real time, providing guidance for early intervention and prevention. Based on the geographical environment stroke trigger trend data, the system will continuously update the dynamic stroke prevention and warning map and generate real-time risk assessment result responses. Through in-depth analysis of risk assessment data, the system can identify the patient's current health risks and predict their development trends. Furthermore, using the preset medical resource matching model, the assessment results are fed back to the referral agency for automated referral push. This mechanism can provide patients with timely medical referral recommendations when the risk of stroke is high, ensuring that patients can quickly obtain professional treatment when needed, and effectively reducing the lethality and serious consequences after a stroke occurs. Therefore, the present invention optimizes the traditional method of constructing a stroke prevention map system based on artificial intelligence, solves the problems of low accuracy in stroke risk prediction and identification, and inability to accurately warn of risk trends in the traditional method of constructing a stroke prevention map system based on artificial intelligence, improves the accuracy of stroke risk prediction and identification, and enhances the ability to warn of risk trends.

[0006] Preferably, step S1 comprises the following steps: Step S11: collecting real-time physiological sign data of stroke patients through wearable devices to obtain real-time physiological sign data of stroke patients; Step S12: collecting historical case data of stroke patients to obtain historical case data of stroke patients; Step S13: desensitizing the real-time physiological sign data of the stroke patients and the historical case data of the stroke patients, respectively, to obtain desensitized real-time physiological sign data of the stroke patients and desensitized historical case data of the stroke patients; Step S14: Fill missing values ​​in the desensitized real-time vital sign data of stroke patients to obtain real-time vital sign filled data of stroke patients.

[0007] The present invention collects the physiological sign data of stroke patients in real time through wearable devices, including but not limited to key indicators such as heart rate, blood pressure, body temperature, and oxygen saturation. These data provide timely and continuous health monitoring for doctors and medical teams, and can reflect the real-time health status of patients, thereby providing a basis for judging stroke risks and implementing preventive measures. In addition, real-time data collection also makes patient health management more accurate, timely identifies potential stroke danger signals, and conducts early intervention. By collecting historical case data of stroke patients, the medical team can fully understand the patient's previous health status, including past stroke attacks, chronic disease history, drug use, lifestyle, etc. The collection of these data helps to form a complete health record of the patient, providing important support for subsequent risk assessment and personalized customization of treatment plans. Historical case data can reveal potential chronic diseases and other causes of stroke, and provide strong data support for timely intervention and prevention. Data desensitization is a key link in protecting patient privacy and personal information security. After desensitization of the real-time physiological sign data and historical case data of stroke patients, sensitive information (such as name, ID number, contact information, etc.) is removed or anonymized, thereby avoiding the risk of data leakage. The desensitized data still retains sufficient medical value and can be used for clinical analysis, research, and model training without infringing on the privacy rights of patients. This step ensures the compliance of data use while also protecting the personal information security of patients. Missing values ​​are filled in for real-time vital sign data of stroke patients to ensure the integrity and accuracy of the data. Since sensor failure, signal loss, or temporary inability to obtain certain vital sign data may occur during the data collection process, the filling of missing values ​​can be supplemented by a variety of methods (such as interpolation, mean filling, etc.), making the data set more complete and continuous, thereby avoiding analysis bias caused by missing data. Through this step, the processed real-time vital sign data can be used as the basis for subsequent analysis and prediction models, improving the reliability and accuracy of the overall analysis results.

[0008] Preferably, step S2 comprises the following steps: Step S21: extracting the patient's environment from the real-time vital sign filling data of the stroke patient to obtain the patient's environment data; Step S22: performing a cardiovascular environmental impact assessment on the real-time vital sign filled data of the stroke patient in the exercise state according to the patient's environment data, to obtain cardiovascular environmental impact data in the exercise state; Step S23: performing case triggering cause analysis on the desensitized historical case data of stroke patients to obtain triggering cause data of stroke patients; Step S24: calculating the probability of sudden stroke risk for the stroke patient's triggering cause data according to the cardiovascular environment influence data in the exercise state, and obtaining the sudden stroke risk probability data; Step S25: Calculate the local risk environment level of stroke for the patient's environmental data based on the stroke sudden risk probability data and the cardiovascular environmental impact data under exercise status to obtain the local risk environment level of stroke.

[0009] The present invention extracts the environmental data of the patient, such as temperature, humidity, air pressure, air quality, noise level, altitude and other external factors, by analyzing the real-time vital sign data of the stroke patient. These environmental factors can have a direct impact on the health status of the stroke patient, especially when the external environment changes greatly, it will increase the risk of stroke or trigger a stroke attack. Extracting environmental data helps to further understand the patient's living and activity environment, thereby providing more dimensional information for risk assessment, and helps to achieve personalized risk prediction and intervention. The cardiovascular system of the stroke patient in the state of exercise, especially during physical activity or exercise, is evaluated. In the state of exercise, the patient's cardiovascular system is more sensitive to the external environment (such as temperature, humidity, air quality, etc.), and certain environmental factors have an adverse effect on the heart health of the patient in exercise. This step quantifies the interaction between exercise and environment by analyzing these effects, and obtains the cardiovascular environmental impact data under the state of exercise. These data provide the necessary environmental background for the subsequent prediction and prevention of sudden stroke risk, and help to identify environmental factors that increase the risk of stroke during exercise. Based on the desensitized data of historical cases of stroke patients, the trigger factors are analyzed to identify the potential triggers that cause stroke in patients. These triggers can include emotional stress, eating habits, drug use, sleep quality, lifestyle and other factors. Through in-depth mining of historical data, the system can discover the key factors that affect the occurrence of stroke and evaluate their impact on the risk of stroke based on the frequency and severity of these triggers. Trigger factor analysis not only provides a basis for the design of personalized treatment plans, but also helps predict the risk of stroke that patients face in the future. According to the cardiovascular environmental impact data under exercise and the patient's trigger factor data, the probability of sudden stroke risk is calculated. The environmental impact under exercise interacts with multiple factors such as the patient's physical activity level and cardiovascular health status to affect the possibility of stroke. At the same time, trigger factors (such as high blood pressure, strenuous exercise, etc.) also increase the risk of stroke. By combining these factors, the system can accurately calculate the probability of sudden stroke risk for each patient in a specific environment and state, thereby helping to formulate effective prevention strategies. Based on the stroke sudden risk probability data and the cardiovascular environmental impact data under exercise, the system calculates the local risk environment level of stroke in the patient's environment. By comprehensively considering the patient's physiological state, environmental factors and the probability of sudden risk, the system can classify different risk levels for the patient's region or living environment. This calculation helps to identify which areas or environments pose a higher risk to stroke patients, thereby providing decision support for the medical team to take appropriate preventive measures or conduct closer monitoring to ensure the safety and health of patients.

[0010] Preferably, step S22 includes the following steps: Step S221: analyzing the heart rate fluctuation of the stroke patient in the exercise state on the real-time vital sign filling data to obtain the heart rate fluctuation data of the stroke patient in the exercise state; Step S222: extracting the temperature, humidity and altitude change data of the patient's environment to obtain the temperature and humidity change data of the patient's environment and the altitude change data of the patient; Step S223: quantifying the rapid breathing and heart rate of the stroke patient's exercise state heart rate fluctuation data according to the patient's altitude change data to obtain the rapid breathing and heart rate quantification data; Step S224: performing a temperature regulation heart rate load increment simulation on the heart rate fluctuation data of the stroke patient in motion based on the temperature and humidity change data of the patient's environment to obtain the temperature regulation heart rate load increment data; Step S225: Evaluate the cardiovascular environmental impact under exercise conditions based on the rapid breathing and heart rate quantification data and the temperature regulation heart rate load increment data to obtain cardiovascular environmental impact data under exercise conditions.

[0011] The real-time physical sign data of the stroke patient of the present invention is analyzed, and the heart rate fluctuation under the exercise state is paid attention to. This analysis evaluates the stability of the patient's cardiovascular system by detecting the fluctuation pattern of the heart rate during exercise. The heart rate fluctuation under the exercise state can reveal the changes in the patient's heart burden under different exercise intensities. Especially for stroke patients, excessive heart rate fluctuation is a potential risk indicator of stroke recurrence. Therefore, this step helps to identify the patient's exercise load and heart health status, and provides data support for the formulation of personalized exercise programs and stroke prevention measures. Analyze the environmental data of the stroke patient, especially the changes in temperature and humidity and altitude. These environmental factors have a significant impact on the patient's physiological state, especially in high temperature, high humidity or high altitude areas, the physiological burden of the human body will increase, leading to cardiovascular stress response. The extraction of temperature, humidity and altitude changes provides an environmental background for subsequent analysis, which can help the medical team identify environmental conditions that increase the risk of stroke, so as to intervene and prevent the patient's health management more accurately. Based on the altitude change data of the patient, the heart rate fluctuation of the stroke patient's exercise state is quantified by the rapid breathing and heart rate. The increase in altitude usually leads to a decrease in oxygen content, which causes shortness of breath, which in turn affects the fluctuation of heart rate. Especially for stroke patients, shortness of breath is a sign of increased cardiovascular burden, further increasing the risk of stroke. Therefore, by quantifying the rapid breathing and heart rate, doctors can understand the physiological adaptation of patients under different altitude conditions, and conduct risk assessment and personalized intervention. Based on the temperature and humidity change data, the temperature regulation heart rate load increment simulation is performed on the heart rate fluctuations in the exercise state of stroke patients. Changes in temperature and humidity affect the body's ability to regulate temperature. Especially in high temperature or high humidity environments, the patient's heart will bear more load in the process of regulating body temperature, resulting in an increase in heart rate. The simulation of the temperature regulation heart rate load increment can reflect the specific impact of environmental temperature and humidity changes on the patient's heart burden, and provide important data for evaluating the physiological response of stroke patients under different climatic conditions, thereby helping the medical team to develop a more reasonable treatment plan. Combined with the quantitative data of rapid breathing and heart rate and the temperature regulation heart rate load increment data, the cardiovascular environmental impact assessment under exercise is carried out. This comprehensive assessment takes into account the impact of the patient's exercise state on the cardiovascular system under different environmental conditions (such as temperature and humidity changes, altitude, etc.). Through this analysis, it is possible to quantify how environmental factors affect the cardiac burden and stroke risk of stroke patients, provide patients with cardiovascular health assessments in specific environments, and help doctors make more accurate risk predictions and personalized interventions, thereby improving long-term health management and prevention measures for stroke patients.

[0012] Preferably, step S223 includes the following steps: The altitude difference per unit time is calculated for the patient's altitude change data using a preset time step to obtain the altitude difference per unit time; Evaluate the air thinness degree of the altitude difference per unit time to obtain air thinness degree data; According to the data of air thinness, the slowness of air pressure adaptation is analyzed on the heart rate fluctuation data of stroke patients in motion state, and the slowness of air pressure adaptation of heart rate state is obtained; Based on the data of air thinness and the slowness of heart rate adaptation to air pressure, the short-term altitude emergency response intensity was simulated on the heart rate fluctuation data of stroke patients in motion, and the short-term altitude emergency response intensity data was obtained. According to the short-term altitude emergency reaction intensity data and the heart rate state air pressure adaptation slowness data, the heart rate fluctuation data of the stroke patients in the exercise state are used to quantify the rapid breathing and heart rate, and the rapid breathing and heart rate quantitative data are obtained.

[0013] The present invention uses a preset time step to calculate the altitude difference per unit time for the altitude change data of the patient. The change in altitude affects the patient's physiological response, especially in a rapidly changing environment, when the altitude changes greatly, the oxygen concentration and air pressure change more significantly. By accurately calculating the altitude difference per unit time, the impact of these environmental factors on the patient's physiological state can be quantified, and basic data is provided for further analysis of the impact of altitude changes on the cardiovascular system. This step can help understand the degree of environmental stress faced by the patient, especially for patients with a history of stroke, rapid altitude changes increase the burden on the heart and respiratory system. By evaluating the air thinness of the altitude difference per unit time, the air thinness data is obtained. With the increase in altitude, the oxygen concentration in the air gradually decreases, which will affect various physiological functions of the human body, especially in high altitude areas, the air thinness leads to an increase in the load on the heart and respiratory system. By evaluating the air thinness, the specific impact of insufficient oxygen supply on the patient's health under specific altitude conditions can be clarified, especially for stroke patients, this information is crucial to the formulation of personalized health management plans, and helps identify potential risks in the environment. According to the air thinness data, in the third sub-step, the heart rate fluctuation data of the stroke patient in the state of exercise is analyzed for air pressure adaptation slowness, and the heart rate state air pressure adaptation slowness data is obtained. Air pressure changes, especially in rapid altitude changes or high altitude environments, can affect the patient's cardiovascular system response, resulting in changes or instability in heart rate. The air pressure adaptation slowness analysis aims to evaluate the patient's heart's ability to adapt to air pressure changes, that is, the degree of delay in the heart rate's response to air pressure changes. For stroke patients, slow heart adaptation leads to a high risk of cardiovascular events. Analyzing this indicator can help predict increased cardiac load and provide a reference for personalized treatment plans. By simulating the intensity of short-term altitude emergency response to the heart rate fluctuation data of the stroke patient in the state of exercise based on the air thinness data and the heart rate state air pressure adaptation slowness data, the short-term altitude emergency response intensity data is obtained. Rapid changes in altitude will prompt the body to respond in a short time, such as speeding up the heartbeat to compensate for insufficient oxygen. This simulation can quantify the intensity of the patient's emergency response when facing a short-term altitude change. Especially for stroke patients, strong emergency responses lead to excessive heart burden and a surge in the risk of stroke. Therefore, this simulation helps to assess the health risks faced by patients in emergency situations, thereby helping to take preventive measures in advance. Based on the short-term altitude emergency response intensity data and the heart rate state pressure adaptation slowness data, the heart rate fluctuation data of stroke patients in exercise status are quantified to obtain the rapid breathing and heart rate quantitative data. Altitude changes not only affect heart rate fluctuations, but also cause shortness of breath, especially in an oxygen-deficient environment, the body tries to compensate for the lack of oxygen by speeding up breathing.The quantification of shortness of breath is a detailed analysis of the patient's respiratory system response, which can reveal whether the patient's respiratory system is overloaded under the influence of exercise and altitude changes. For stroke patients, shortness of breath is an important clinical indicator. Through this quantitative analysis, potential cardiovascular and respiratory health risks can be identified, and intervention can be carried out in advance to reduce the possibility of sudden stroke.

[0014] Preferably, step S224 includes the following steps: Calculate the temperature and humidity variation coefficient for the temperature and humidity variation data of the patient's environment to obtain the temperature and humidity variation coefficient; The body temperature heat stress response is simulated based on the temperature and humidity variation coefficients of the heart rate fluctuation data of the stroke patients in motion state, and the body temperature heat stress response simulation data is obtained; Conducting thermoregulatory effect coupling on the body temperature heat stress response simulation data to obtain heat stress thermoregulatory effect data; According to the thermoregulatory effect data of heat stress, the thermoregulatory heart rate load increment data of the stroke patients in motion state were simulated to obtain the thermoregulatory heart rate load increment data.

[0015] The present invention uses the temperature and humidity change data of the patient's environment to calculate the temperature and humidity change amplitude coefficient to obtain the temperature and humidity change amplitude coefficient. The fluctuation of environmental temperature and humidity directly affects the physiological regulation function of the human body, especially in the case of drastic temperature changes or high humidity, the temperature regulation mechanism in the patient's body is challenged. By calculating the temperature and humidity change amplitude coefficient, the temperature and humidity change amplitude of the environment can be quantified, so as to evaluate the degree of influence of the environment on the patient's body temperature regulation. For stroke patients, the impact of temperature and humidity fluctuations is more significant. This step provides basic data for subsequent heat stress simulation and helps identify the health risks brought by environmental factors. According to the temperature and humidity change amplitude coefficient, the second sub-step simulates the body temperature heat stress response of the stroke patient's exercise state heart rate fluctuation data. The purpose of the heat stress response simulation is to evaluate the patient's body temperature regulation mechanism and heart rate response under the change of environmental temperature and humidity. The drastic change of environmental temperature and humidity will trigger a heat stress response, which will cause the heart and vascular system to bear an additional burden. Especially for stroke patients, heat stress will increase cardiovascular risk and even cause stroke recurrence or other complications. Through this simulation, the pressure of environmental changes on the patient's heart can be evaluated, and a basis for subsequent treatment strategies can be provided. The thermoregulatory effect is coupled to the simulated data of the body temperature heat stress response to obtain the thermoregulatory effect data of heat stress. The body's thermoregulatory mechanisms, such as heat dissipation, sweating, and vasoconstriction, are automatically activated in the face of heat stress to keep the body temperature stable. However, these regulatory processes require additional energy consumption, especially when the ambient temperature and humidity are high, the regulatory effect is limited. By coupling the thermoregulatory effect, it is possible to more accurately evaluate how stroke patients adapt to and cope with heat stress in a high temperature or humidity environment, and whether this adaptation is sufficient to avoid potential health risks. The results of this step provide an important reference for subsequent heart rate load simulation. Based on the thermoregulatory effect data of heat stress, the thermoregulatory heart rate load increment is simulated for the heart rate fluctuation data of stroke patients in the exercise state, and the thermoregulatory heart rate load increment data is obtained. The thermoregulatory burden caused by changes in temperature and humidity directly affects the patient's heart rate changes, especially under extreme environmental conditions, the heart must make more efforts to support thermoregulation. Through this simulation, the increment of the heart load caused by the thermoregulatory process can be quantified, especially for stroke patients, this heart rate load increment will have a significant impact on their heart health. The simulation results provide strong support for formulating targeted health management measures, assessing risks and predicting cardiovascular events, which helps to optimize treatment plans and reduce the health risks of stroke patients in special environments.

[0016] Preferably, step S24 comprises the following steps: Step S241: performing triggering environment association with the triggering cause data of the stroke patient according to the cardiovascular environment influence data in the exercise state to obtain triggering cause environment association data; Step S242: performing abnormal blood dynamics simulation for stroke based on the cardiovascular environment influence data and the cause-triggered environment association data in the exercise state to obtain abnormal blood dynamics simulation data for stroke; Step S243: identifying environmental factor weighted risk factors for cardiovascular environmental impact data under exercise status according to the inducement triggering environment association data and stroke blood abnormality dynamic simulation data, and obtaining environmental factor weighted risk factors; Step S244: Calculate the probability of sudden stroke risk based on the triggering cause data of the stroke patient based on the weighted risk factor of the environmental factors, the abnormal blood dynamics simulation data of the stroke, and the triggering cause environment association data to obtain the sudden stroke risk probability data.

[0017] The present invention associates the triggering inducement data of stroke patients with the triggering inducement environment according to the cardiovascular environmental influence data under the motion state, and obtains the triggering inducement environment association data. The cardiovascular environmental influence data under the motion state include factors such as temperature and humidity, air quality, and air pressure changes, which have a direct impact on the patient's cardiovascular system. The triggering inducement data of stroke patients include excessive exercise, emotional fluctuations, drastic changes in the environment, etc. By associating these data, the relationship between specific environmental factors and inducements can be identified, thereby revealing which environmental conditions will trigger the occurrence of stroke or aggravate stroke symptoms. The generation of this data helps to identify the risk environment and further provide support for personalized risk management and prevention. Based on the cardiovascular environmental influence data under the motion state and the inducement triggering environment association data, the abnormal dynamics of stroke blood are simulated to obtain the abnormal dynamics of stroke blood simulation data. Exercise and environmental factors have a profound impact on blood flow and the physical and chemical properties of blood, especially in stroke patients, indicators such as blood viscosity, hemodynamic changes, and oxygen transport efficiency will be disturbed. The simulation of abnormal blood dynamics can help analyze how changes in blood rheology exacerbate the pathological process of stroke under different motion states and environmental influences. The simulation results can provide doctors with early warnings about the risk of stroke recurrence and aggravation, ensuring accurate control of the patient's condition. According to the environmental association data of triggering factors and the simulation data of abnormal blood dynamics of stroke, the weighted risk factors of environmental factors are identified for the cardiovascular environmental impact data under motion state, and the weighted risk factors of environmental factors are obtained. The core purpose of this step is to identify the most critical risk factors from the complex relationship between environmental data, hemodynamic changes and triggering factors. These risk factors include temperature changes, humidity changes, air pressure fluctuations, etc., as well as their degree of influence on the cardiovascular system under different circumstances. Through the identification of weighted risk factors, it can be revealed which environmental factors have the greatest impact on the occurrence of stroke, thereby providing a strong basis for the optimization of prevention and treatment plans. Based on the weighted risk factors of environmental factors, the simulation data of abnormal blood dynamics of stroke and the environmental association data of triggering factors, the stroke patient triggering triggering data is used to calculate the risk probability of sudden stroke, and the risk probability data of sudden stroke is obtained. This step comprehensively analyzes various variables to predict the risk probability of sudden stroke. By combining hemodynamic changes, environmental risk factors, and specific scenarios of triggers, the probability of stroke is calculated. This probability calculation provides doctors with a quantitative risk assessment tool to help identify potential stroke risks in advance and develop personalized prevention and intervention measures.

[0018] The beneficial effects of the present invention are as follows: the present invention optimizes a traditional method for constructing a stroke prevention map system based on artificial intelligence, solves the problems of low accuracy in stroke risk prediction and identification, and inability to accurately warn of risk trends in the traditional method for constructing a stroke prevention map system based on artificial intelligence, improves the accuracy of stroke risk prediction and identification, and enhances the ability to warn of risk trends. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1 A schematic diagram of the steps of a method for constructing a stroke prevention map system based on artificial intelligence; Figure 2 for Figure 1 Detailed implementation steps of step S2 in FIG.

[0020] The realization of the purpose, functional features and advantages of the present invention will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0021] The technical method of the present invention is described clearly and completely below in conjunction with the accompanying drawings. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by technicians in this field without creative work are within the scope of protection of the present invention.

[0022] In addition, the accompanying drawings are only schematic illustrations of the present invention and are not necessarily drawn to scale. The same reference numerals in the figures represent the same or similar parts, and their repeated description will be omitted. Some of the block diagrams shown in the accompanying drawings are functional entities and do not necessarily correspond to physically or logically independent entities. The functional entities can be implemented in software form, or implemented in one or more hardware modules or integrated circuits, or implemented in different networks and / or processor methods and / or microcontroller methods.

[0023] It should be understood that, although the terms "first", "second", etc. may be used herein to describe various units, these units should not be limited by these terms. These terms are used only to distinguish one unit from another unit. For example, without departing from the scope of the exemplary embodiments, the first unit may be referred to as the second unit, and similarly the second unit may be referred to as the first unit. The term "and / or" used herein includes any and all combinations of one or more of the listed associated items.

[0024] To achieve this, please refer to Figure 1 to Figure 2 , a method for constructing a stroke prevention map system based on artificial intelligence, the method comprising the following steps: Step S1: collecting real-time physiological sign data of stroke patients through wearable devices to obtain real-time sign filling data of stroke patients; collecting historical case data of stroke patients to obtain desensitized historical case data of stroke patients; Step S2: extracting the patient's environment from the real-time vital sign filling data of the stroke patient to obtain the patient's environment data; analyzing the case triggering factors of the stroke patient's historical case desensitization data to obtain the stroke patient's triggering factor data; calculating the stroke sudden risk probability of the stroke patient's triggering factor data to obtain the stroke sudden risk probability data; calculating the stroke local risk environment level of the patient's environment data based on the stroke sudden risk probability data to obtain the stroke local risk environment level; Step S3: performing a geographical environment stroke triggering trend analysis on the stroke sudden risk probability data according to the stroke local risk environment level to obtain geographical environment stroke triggering trend data; constructing a dynamic stroke prevention and early warning map based on the geographical environment stroke triggering trend data to obtain a dynamic stroke prevention and early warning map; Step S4: respond to the risk assessment results of the dynamic stroke prevention and early warning map based on the stroke triggering trend data of the geographical environment to obtain risk assessment result response data, and use the preset medical resource matching model to push the risk assessment result response data to the referral agency to obtain stroke risk referral agency push data.

[0025] In the embodiment of the present invention, reference Figure 1 The above is a schematic flow chart of the steps of a method for constructing a stroke prevention map system based on artificial intelligence of the present invention. In this example, the method for constructing a stroke prevention map system based on artificial intelligence includes the following steps: Step S1: collecting real-time physiological sign data of stroke patients through wearable devices to obtain real-time sign filling data of stroke patients; collecting historical case data of stroke patients to obtain desensitized historical case data of stroke patients; In the embodiment of the present invention, firstly, real-time physiological sign data of stroke patients are collected through wearable devices, including but not limited to physiological indicators such as heart rate, blood pressure, blood oxygen saturation, body temperature, and respiratory rate. The wearable device uses sensors to monitor the changes of the patient's physical signs in real time, and transmits the data to the central processing unit through wireless transmission. In order to ensure the integrity and accuracy of the collected data, all transmitted data will first be encrypted to ensure data security. The collected physiological sign data are interpolated and smoothed by the data processing algorithm to solve the problem of data missing caused by unstable signals or transmission interruptions, and finally generate complete real-time physical sign filling data of stroke patients. At the same time, the system will also collect historical case data of stroke patients. Historical case data includes but is not limited to the patient's past medical history, drug use, previous disease information, etc., which will be extracted through the hospital information management system. The historical case data will be desensitized during the collection process, that is, the patient's name, ID number and other personal identity information will be de-identified to ensure data privacy protection. During the desensitization process, a standard anonymization algorithm is used to ensure that the data can be used for analysis, but the patient's personal information will not be leaked. The anonymized historical case data will be used for subsequent trigger analysis to generate anonymized data of stroke patients' historical cases. Finally, after completing this step, the two sets of data obtained are the real-time vital sign filling data of stroke patients and the anonymized data of stroke patients' historical cases, which will be used for risk assessment and analysis in subsequent steps.

[0026] Step S2: extracting the patient's environment from the real-time vital sign filling data of the stroke patient to obtain the patient's environment data; analyzing the case triggering factors of the stroke patient's historical case desensitization data to obtain the stroke patient's triggering factor data; calculating the stroke sudden risk probability of the stroke patient's triggering factor data to obtain the stroke sudden risk probability data; calculating the stroke local risk environment level of the patient's environment data based on the stroke sudden risk probability data to obtain the stroke local risk environment level; In the embodiment of the present invention, in step S2, the real-time vital sign filling data of the stroke patient is firstly used to extract the patient's environment. By combining sensor equipment (such as temperature and humidity sensors, air pressure sensors, etc.), the system can obtain the patient's environmental conditions in real time. The environmental data collected by the sensor, including factors such as temperature, humidity, air quality, and air pressure, is combined with GPS positioning information to accurately determine the specific geographical location of the patient. Through the environmental data acquisition system, the patient's environmental data is obtained, and these data are standardized to eliminate errors caused by factors such as equipment differences and environmental changes, thereby ensuring the accuracy and consistency of the environmental data. Secondly, the system performs case triggering cause analysis on the historical case desensitization data of the stroke patient. In this process, the logistic regression analysis method is used to analyze which causes will cause the sudden risk of stroke based on the patient's past disease history, living habits, and other factors. For example, the relationship between the patient's history of hypertension, diabetes, heart disease, and other factors and the onset of stroke, and the corresponding triggering cause is extracted through the model. The analysis results will generate the triggering cause data of the stroke patient. Then, based on the triggering cause data of the stroke patient, the system calculates the probability of sudden stroke risk. In this process, the Bayesian risk analysis method is used. By establishing a conditional probability model between triggering factors and stroke occurrence, the probability of a patient having a stroke in the current environment and physiological state is calculated. The risk probability calculation takes into account the comprehensive influence of multiple variables such as the patient's physiological state, previous medical history, and environmental factors. Finally, the stroke sudden risk probability data is obtained. On this basis, based on the stroke sudden risk probability data, the system calculates the local risk environment level of the patient's environmental data. To this end, the analytic hierarchy process (AHP) is used for comprehensive evaluation. By weighting different environmental factors (such as temperature, humidity, air quality, etc.), the importance of each environmental factor in the occurrence of stroke risk is calculated, and the local risk environment level of stroke is finally calculated by combining the patient's real-time physiological data with the sudden risk probability. This level reflects the potential stroke risk level of the patient's environment. Finally, the local risk environment level data of stroke is obtained.

[0027] Step S3: performing a geographical environment stroke triggering trend analysis on the stroke sudden risk probability data according to the stroke local risk environment level to obtain geographical environment stroke triggering trend data; constructing a dynamic stroke prevention and early warning map based on the geographical environment stroke triggering trend data to obtain a dynamic stroke prevention and early warning map; In the embodiment of the present invention, first, according to the local risk environment level of stroke, the patient's environmental data is sorted in descending order of risk point distribution. This process uses a sorting algorithm to arrange all monitored environmental data points (such as temperature and humidity, air pressure, pollutant concentration, etc.) from high to low according to the risk level. The sorting standard is based on the local risk environment level of stroke calculated in the previous step, and environmental data points with higher risk levels are prioritized. In the sorting process, a quick sorting algorithm is used to improve processing efficiency and ensure that a large amount of data can be sorted in a short time. Next, based on the sorted risk point data, the system performs a geographical environment stroke triggering trend analysis on the stroke sudden risk probability data. This process first models the relationship between each risk point in the geographical area through a spatial interpolation method, uses the nearest neighbor algorithm to evaluate the risk impact of the surrounding environment, and predicts the change trend of the risk point through time series analysis. The geographical environment stroke triggering trend analysis helps to reveal how different environmental factors affect the temporal and spatial distribution of stroke sudden risk. Finally, based on the geographical environment stroke triggering trend data, the system constructs a dynamic stroke prevention and warning map. During this process, the system visualizes all the analyzed environmental data, risk levels, sudden risk probability and other information in the form of a map. In order to ensure that the map can reflect environmental changes in real time, the system uses dynamic data stream processing technology to ensure that the warning map can be automatically refreshed as the data is updated. Through this dynamic stroke prevention warning map, medical staff can promptly grasp the high-risk areas for stroke and take effective preventive measures.

[0028] Step S4: respond to the risk assessment results of the dynamic stroke prevention and early warning map based on the stroke triggering trend data of the geographical environment to obtain risk assessment result response data, and use the preset medical resource matching model to push the risk assessment result response data to the referral agency to obtain stroke risk referral agency push data.

[0029] In an embodiment of the present invention, based on the geographical environment stroke triggering trend data, the system responds to the risk assessment results. The process first compares the geographical environment stroke triggering trend data with the known stroke occurrence history to evaluate the sudden risk of each geographical area. Through the multi-factor decision analysis method, the system evaluates the correlation between different environmental factors and stroke suddenness, and combines the stroke sudden risk probability data to generate corresponding risk assessment result response data for each area. Using the preset medical resource matching model, the system pushes the risk assessment result response data to the referral agency. First, by analyzing the medical resources of the referral agency (such as the number of beds, professional doctors, equipment, etc.), the system matches a suitable medical institution for each high-risk area. Then, based on the patient's real-time physiological signs, geographical location and other information, the system automatically pushes the nearest referral agency information to ensure that the patient can receive medical intervention in the shortest time and reduce the mortality or disability rate after the stroke occurs. Finally, the stroke risk referral agency push data is obtained, and medical staff and relevant departments can take intervention measures in time according to the data to improve the efficiency and accuracy of stroke prevention.

[0030] Preferably, step S1 comprises the following steps: Step S11: collecting real-time physiological sign data of stroke patients through wearable devices to obtain real-time physiological sign data of stroke patients; Step S12: collecting historical case data of stroke patients to obtain historical case data of stroke patients; Step S13: desensitizing the real-time physiological sign data of the stroke patients and the historical case data of the stroke patients, respectively, to obtain desensitized real-time physiological sign data of the stroke patients and desensitized historical case data of the stroke patients; Step S14: Fill missing values ​​in the desensitized real-time vital sign data of stroke patients to obtain real-time vital sign filled data of stroke patients.

[0031] In an embodiment of the present invention, real-time physiological sign data of stroke patients are collected through wearable devices. Specifically, the sensors built into the device include a heart rate sensor, a blood pressure sensor, a blood oxygen saturation sensor, a body temperature sensor, and a respiratory rate sensor. The sensor continuously collects the patient's physiological data and transmits the collected data to the data receiving center in real time through a wireless communication module (such as Bluetooth or Wi-Fi). The data of each sensor will undergo a preliminary quality check to ensure data integrity. If the sensor fails to work properly, the system will automatically trigger an alarm mechanism so that it can be replaced or repaired in time. The transmitted data format adopts a unified standard, such as JSON format, to ensure the compatibility and parsability of the data. All physiological sign data will be recorded in the form of a timestamp to ensure the accuracy of its timing. During the data transmission process, in order to ensure the privacy and security of the data, all data are encrypted and encrypted using the AES encryption algorithm. Once the data arrives at the data receiving center, it will be stored in the database and marked as "real-time physiological sign data". Finally, the result is real-time physiological sign data of stroke patients, which includes but is not limited to key indicators such as the patient's heart rate, blood pressure, blood oxygen saturation, and body temperature. The historical case data of stroke patients is collected. Historical case data include the patient's medical history, medication history, allergy history, family history, and past examination reports and medical records. Data collection depends on the hospital's information management system, which extracts the patient's historical case data by connecting to the hospital's electronic health record system (EHR). During the collection process, the system automatically checks the integrity of the data to ensure that each historical data accurately reflects the patient's medical background. For example, the examination report should include relevant information such as imaging examinations, laboratory test results, and previous stroke attacks. In order to avoid human errors, the data collection process is carried out through automated scripts and programs without manual intervention, thereby improving efficiency and accuracy. During data processing, the hospital information management system will anonymize the patient's personal information, that is, delete all sensitive data that reveals personal identity (such as name, ID number, etc.). The technology used for data desensitization is hash encryption, that is, the patient's sensitive information is converted into an irreversible hash value through an encryption algorithm, thereby effectively protecting privacy. In the end, the result is the desensitized historical case data of stroke patients, which is saved as a structured database file for subsequent processing. In step S13, the real-time physiological sign data and historical case data of stroke patients are desensitized. First, for real-time physiological sign data, when desensitizing the data, the part related to the patient's personal identity is mainly protected. This includes the patient's name, identity information, place of residence, contact information, etc. The desensitization process converts sensitive data fields (such as name, ID number, etc.) using encryption algorithms to ensure that sensitive data cannot be restored.For each data point in the real-time vital signs data, in addition to physiological indicators, other personal information will be processed similarly. The system anonymizes the physiological data of each patient through an identifier, such as using "patient ID" as the unique identifier of each patient, so that all physiological data can be associated while ensuring privacy protection. For the desensitization of historical case data, it is also necessary to remove the patient's identity information and hide sensitive information such as drug use history and medical records. Here, data masking technology is used to set desensitization rules to shield specific drug names, disease details, and doctor's diagnosis information, leaving only fuzzy information or category information. After desensitization, historical case data can still reflect the patient's medical record information, but no specific personal identity can be restored. Finally, after desensitization, two sets of data are obtained: real-time vital signs desensitized data of stroke patients and historical cases of stroke patients. These data can be used for subsequent analysis and calculation. The location of missing values ​​is identified through simple judgment rules. If the data of a sensor is missing within a certain time interval, the data at that time point is marked as missing. Different filling methods are used according to the characteristics of different data types. For continuous data (such as blood pressure, body temperature, etc.), linear interpolation is used to fill missing values. Specifically, the missing data is estimated by the average value of the valid data points before and after. For example, if the blood pressure value at a certain time point is missing, the missing value at that position is filled by calculating its linear average value through the blood pressure data of the previous and next time points. During the filling process, the system will continuously check whether the filled data meets the consistency requirements of the previous and next data. If the filled data exceeds the normal range (such as abnormal blood pressure values), the alarm mechanism will be triggered for manual intervention. After filling the missing values, the system will perform time series smoothing on the data to ensure the stability of the data and avoid abnormal fluctuations caused by filling. The sliding window algorithm is used here to process the data, that is, by locally smoothing the filled data, it is ensured that the data change trend is consistent with the actual physiological state.

[0032] Preferably, step S2 comprises the following steps: Step S21: extracting the patient's environment from the real-time vital sign filling data of the stroke patient to obtain the patient's environment data; Step S22: performing a cardiovascular environmental impact assessment on the real-time vital sign filled data of the stroke patient in the exercise state according to the patient's environment data, to obtain cardiovascular environmental impact data in the exercise state; Step S23: performing case triggering cause analysis on the desensitized historical case data of stroke patients to obtain triggering cause data of stroke patients; Step S24: calculating the probability of sudden stroke risk for the stroke patient's triggering cause data according to the cardiovascular environment influence data in the exercise state, and obtaining the sudden stroke risk probability data; Step S25: Calculate the local risk environment level of stroke for the patient's environmental data based on the stroke sudden risk probability data and the cardiovascular environmental impact data under exercise status to obtain the local risk environment level of stroke.

[0033] As an example of the present invention, refer to Figure 2 As shown, in this example, step S2 includes: Step S21: extracting the patient's environment from the real-time vital sign filling data of the stroke patient to obtain the patient's environment data; In an embodiment of the present invention, the real-time vital sign filling data of the stroke patient is used to extract the patient's environment. The patient's environmental data includes environmental factors such as temperature, humidity, air pressure, air quality, noise level, light intensity, altitude, etc. These environmental data can be collected in real time by environmental monitoring equipment (such as environmental sensors, weather stations, etc.), and synchronized with the patient's real-time vital sign data through a data acquisition system. The environmental data collector is installed in the patient's living or activity area, and the collected data is transmitted to the central database through wireless transmission technology. In order to ensure the accuracy of the data, the environmental sensor is calibrated during installation and regularly maintained to ensure that the data is correct. Each environmental data point includes its acquisition timestamp to ensure the time alignment of the data. In the environmental data, data such as temperature and humidity are stored as time series data for subsequent correlation analysis with the patient's physiological vital sign data. All environmental data and vital sign data are merged and matched through a unified identifier, that is, "patient ID" and "environmental data timestamp" are used as key fields to ensure that the environmental data corresponds correctly to the patient's vital sign data. Ultimately, the result is "patient environment data", which not only reflects the patient's physical environment, but also provides important external factor information for subsequent analysis.

[0034] Step S22: performing a cardiovascular environmental impact assessment on the real-time vital sign filled data of the stroke patient in the exercise state according to the patient's environment data, to obtain cardiovascular environmental impact data in the exercise state; In an embodiment of the present invention, in step S22, the cardiovascular environmental impact assessment of the real-time vital sign filling data of the stroke patient under the motion state is performed based on the environmental data of the patient. Specifically, first, it is necessary to extract the motion state related indicators in the real-time vital sign data of the patient, such as the step frequency, activity intensity, acceleration, posture change and other information. At this time, the motion sensor (such as accelerometer, gyroscope) will record the patient's motion in real time and transmit its data to the data processing system. Then, the multivariate analysis method is used to evaluate the impact of environmental factors on the cardiovascular system under the motion state by combining the patient's motion data with environmental data (such as temperature, humidity, air quality, etc.). The evaluation process includes the following steps: First, analyze the change trend of various physiological signs (such as heart rate, blood pressure, blood oxygen level, etc.) during the patient's exercise. Then, use the weighted regression model to construct an evaluation formula for the impact of the environment on the cardiovascular system according to the weights of different environmental factors. For example, higher temperature and humidity lead to an increased burden on the cardiovascular system, so in the calculation process, it is necessary to correct the changes in the vital sign data according to the changes in environmental factors. The evaluation process is based on statistical methods and regression analysis theory. By combining exercise status and environmental data, the cardiovascular environmental impact data under exercise status is obtained. Ultimately, the output result is the patient's cardiovascular impact score in the current environment, reflecting the pressure of environmental factors on their cardiovascular system.

[0035] Step S23: performing case triggering cause analysis on the desensitized historical case data of stroke patients to obtain triggering cause data of stroke patients; In an embodiment of the present invention, case triggering factor analysis is performed on the desensitized data of historical cases of stroke patients. The purpose of case triggering factor analysis is to find out the triggering factors of stroke events, including mood swings, overwork, dietary problems, improper use of drugs, etc. In order to conduct this analysis, it is first necessary to extract key information related to stroke attacks from the patient's historical case data, such as triggering factors of previous stroke attacks, related symptoms, life events during the attack, etc. At this time, the symptoms and attack scenarios of each record in the case data will be classified by a standardized coding system, such as using ICD-10 codes to encode the disease and mark the specific causes of stroke. Then, the correlation between each inducement and stroke attack is analyzed by association rule analysis. By mining historical case data, frequently occurring inducement combinations are identified. For example, if some patients suffer a stroke after experiencing extreme fatigue or long-term high pressure, these factors are regarded as potential inducements of stroke attacks. During the analysis process, combined with statistical test methods (such as chi-square test), the significance between each inducement and stroke attack is further evaluated. Finally, the generated result is "stroke patient triggering cause data", which is a collection of factors that identify the triggering causes of stroke, which can help in the subsequent analysis of the assessment of sudden stroke risk.

[0036] Step S24: calculating the probability of sudden stroke risk for the stroke patient's triggering cause data according to the cardiovascular environment influence data in the exercise state, and obtaining the sudden stroke risk probability data; In an embodiment of the present invention, the stroke sudden risk probability is calculated based on the triggering inducement data of the stroke patient based on the cardiovascular environmental influence data under the motion state. First, it is necessary to integrate the cardiovascular environmental influence data under the motion state and the triggering inducement data to construct a comprehensive risk assessment model. The specific method is to use a multiple regression analysis model, take the cardiovascular environmental influence data under the motion state and the triggering inducement data as independent variables, and model the stroke risk as the dependent variable. By analyzing the patient's motion state and the influence of the external environment and the inducement that triggers the stroke, the probability of stroke can be predicted. In the calculation process, the weight of each inducement and environmental factor is obtained based on historical cases and experimental data, and the weight value reflects the contribution of different factors to the stroke risk. For example, when some patients perform strenuous exercise in high temperature, low pressure and other environments, the probability of stroke increases, so the weight of these environmental factors will be greater. Through the statistical regression method, a comprehensive risk assessment result, i.e., "stroke sudden risk probability data", is obtained.

[0037] Step S25: Calculate the local risk environment level of stroke for the patient's environmental data based on the stroke sudden risk probability data and the cardiovascular environmental impact data under exercise status to obtain the local risk environment level of stroke.

[0038] In an embodiment of the present invention, based on the stroke sudden risk probability data and the cardiovascular environmental impact data under the motion state, the local risk environment level of the stroke is calculated for the patient's environmental data. First, the stroke sudden risk probability data is combined with the cardiovascular environmental impact data under the motion state to divide the environmental level. According to the stroke sudden risk probability, the system classifies the environment, such as low risk, medium risk, high risk, etc. On this basis, combined with the patient's environmental data (such as air quality, temperature, humidity, etc.), the environmental risk level is further adjusted. For example, in an environment with poor air quality and high temperature, even if the patient's sudden risk probability is low, its environmental risk level should be increased. Through a weighted algorithm, the influence of the sudden risk probability and environmental data is comprehensively considered to obtain the local risk level of the patient's environment. Specifically, a linear weighted model is used to calculate the final environmental level according to the relationship between each environmental factor and the stroke risk. The output of the environmental level will be divided into multiple levels, such as "low risk environment", "medium risk environment", "high risk environment", etc. Finally, the output "stroke local risk environment level" can provide patients with accurate environmental risk prompts and guide patients to make corresponding life and health management decisions.

[0039] Preferably, step S22 includes the following steps: Step S221: analyzing the heart rate fluctuation of the stroke patient in the exercise state on the real-time vital sign filling data to obtain the heart rate fluctuation data of the stroke patient in the exercise state; Step S222: extracting the temperature, humidity and altitude change data of the patient's environment to obtain the temperature and humidity change data of the patient's environment and the altitude change data of the patient; Step S223: quantifying the rapid breathing and heart rate of the stroke patient's exercise state heart rate fluctuation data according to the patient's altitude change data to obtain the rapid breathing and heart rate quantification data; Step S224: performing a temperature regulation heart rate load increment simulation on the heart rate fluctuation data of the stroke patient in motion based on the temperature and humidity change data of the patient's environment to obtain the temperature regulation heart rate load increment data; Step S225: Evaluate the cardiovascular environmental impact under exercise conditions based on the rapid breathing and heart rate quantification data and the temperature regulation heart rate load increment data to obtain cardiovascular environmental impact data under exercise conditions.

[0040] In an embodiment of the present invention, the heart rate fluctuation analysis is performed on the real-time vital sign filling data of the stroke patient under the exercise state. The heart rate fluctuation analysis aims to evaluate the amplitude and frequency characteristics of the heart rate changes of the stroke patient under the exercise state, and then reflect the load of the cardiovascular system. First, the heart rate data is extracted from the real-time vital sign data of the patient, and the timestamp of the data is marked. The heart rate data is filtered by a high-pass filter and a low-pass filter to remove the short-term instantaneous fluctuations and noise from the exercise process to ensure the accuracy of the analysis results. Then, the heart rate data is subjected to fluctuation analysis using time domain and frequency domain analysis methods. In the time domain, the standard deviation (SDNN) and root mean square value (RMSSD) of the heart rate are calculated to reflect the overall fluctuation degree of the heart rate. The frequency domain analysis uses the fast Fourier transform (FFT) algorithm to convert the heart rate fluctuation data into a spectrum diagram, and further evaluate the power values ​​of the low-frequency (LF) and high-frequency (HF) fluctuation intervals. At this time, the LF band reflects the activity of the sympathetic nerves, while the HF band reflects the activity of the parasympathetic nerves. Through the frequency domain analysis of heart rate fluctuations, the balance state of the sympathetic and parasympathetic nerves of stroke patients can be quantitatively evaluated, and the load of their cardiovascular system can be understood. Finally, after filtering, time domain and frequency domain analysis, the obtained motion state heart rate fluctuation data can accurately reflect the cardiovascular system fluctuations generated by stroke patients during exercise. The specific data includes the heart rate fluctuation amplitude, frequency and corresponding physiological response information. In step S222, the temperature, humidity and altitude change data of the patient's environment are extracted to obtain the temperature and humidity change data of the patient's environment and the altitude change data of the patient. First, the temperature and humidity change data are collected by the environmental sensor, which monitors the temperature and humidity of the patient's surrounding environment in real time and records these data regularly. The temperature and humidity changes at continuous time points are calculated by the differential algorithm, and the change amplitude, fluctuation frequency and trend of temperature and humidity are extracted to form temperature and humidity change data. For the altitude change data, the patient's geographical location is obtained by an accurate GPS positioning module, and then the real-time change of altitude is calculated. During the acquisition process, the patient's movement trajectory is recorded by a real-time dynamic tracking system to ensure that the altitude data of the location is accurate. Altitude changes have an important impact on the cardiovascular system, especially oxygen supply. Therefore, changes at different altitudes must be tracked continuously. Finally, the temperature and humidity change data are sorted out with the altitude change data to obtain complete temperature, humidity and altitude change data of the patient's environment, providing reliable environmental data support for the subsequent analysis of the impact on the cardiovascular system. In step S223, the heart rate fluctuation data of the stroke patient in motion is quantified according to the altitude change data of the patient to obtain the rapid breathing and heart rate quantification data. Altitude changes have an important impact on the respiratory function of the human body, especially in a higher altitude environment, the oxygen concentration in the air is reduced, and the patient's breathing frequency and depth will increase.First, based on the altitude change data, the patient's respiratory rate is dynamically monitored and combined with the heart rate data for joint analysis. For each time period of altitude change, the patient's respiratory rate (unit: times / minute) and heart rate (unit: times / minute) are extracted, and the relationship between the two is calculated. Then, the fast Fourier transform (FFT) method is used to perform frequency domain analysis on the respiratory rate to extract the change characteristics of the respiratory rate. By comparing the respiratory rate data under normal altitude environment and the respiratory rate data under high altitude environment, the degree of shortness of breath is quantified, that is, the quantitative data of shortness of breath and heart rate. This data will reflect the changes in respiratory rate caused by the increase in altitude and help evaluate the potential impact of altitude changes on the patient's cardiovascular health. Based on the temperature and humidity change data of the patient's environment, the heart rate fluctuation data of the stroke patient in the state of exercise is simulated for the incremental load of body temperature regulation. First of all, the regulation of temperature and humidity on body temperature has an important physiological effect. In a high temperature and humidity environment, the human body regulates body temperature by sweat evaporation and other means, and this regulation process puts a heavy burden on the cardiovascular system. In order to simulate this load, it is first necessary to calculate the heat stress index of the current environment based on the temperature and humidity change data of the environment. The index can be calculated by the following formula: HI = 0.5 × (T + 61) + ((T − 68) × 1.2) + (RH × 0.094), where HI is the heat stress index, T is the temperature, and RH is the relative humidity. After obtaining the heat stress index, the weighted model can be used to simulate the heart rate load increment during the thermoregulation process. Specifically, as the temperature and humidity increase, the heart rate load of the human body will increase in order to dissipate heat. At this time, the simulation model will calculate the heart rate load increment through regression analysis. Finally, the thermoregulation heart rate load increment data obtained by simulation calculation reflects the increase in the load of the cardiovascular system of stroke patients during exercise due to different environmental temperatures and humidities. The impact of the cardiovascular environment under exercise is evaluated based on the quantitative data of respiratory heart rate and the incremental data of thermoregulation heart rate load. In order to comprehensively evaluate the impact of the environment on the cardiovascular system under exercise, first, the quantitative data of respiratory heart rate and the incremental data of thermoregulation heart rate load are weighted and synthesized. The relative importance of respiratory heart rate and thermoregulation heart rate load in the evaluation is ensured by adjusting the weight value. The multiple linear regression method was used to quantitatively evaluate the impact of the cardiovascular environment during exercise based on the weights obtained from historical data and the actual exercise data of the patient (such as step frequency, acceleration, etc.). In this process, the cardiovascular load assessment model takes into account the patient's physiological response during exercise and dynamically adjusts the cardiovascular load in combination with environmental factors such as temperature, humidity, and altitude changes. Finally, the cardiovascular environmental impact data obtained during exercise reflects the comprehensive impact of environmental factors on the cardiovascular system of stroke patients during exercise, including the pressure on the cardiovascular system caused by factors such as temperature regulation and shortness of breath.

[0041] Preferably, step S223 includes the following steps: The altitude difference per unit time is calculated for the patient's altitude change data using a preset time step to obtain the altitude difference per unit time; Evaluate the air thinness degree of the altitude difference per unit time to obtain air thinness degree data; According to the data of air thinness, the slowness of air pressure adaptation is analyzed on the heart rate fluctuation data of stroke patients in motion state, and the slowness of air pressure adaptation of heart rate state is obtained; Based on the data of air thinness and the slowness of heart rate adaptation to air pressure, the short-term altitude emergency response intensity was simulated on the heart rate fluctuation data of stroke patients in motion, and the short-term altitude emergency response intensity data was obtained. According to the short-term altitude emergency reaction intensity data and the heart rate state air pressure adaptation slowness data, the heart rate fluctuation data of the stroke patients in the exercise state are used to quantify the rapid breathing and heart rate, and the rapid breathing and heart rate quantitative data are obtained.

[0042] In an embodiment of the present invention, first, the altitude data of the patient's location is obtained. The altitude data is usually obtained through a high-precision GPS system to record the altitude changes of the patient over a period of time. By setting a suitable time step (for example, every minute), the altitude value at each time point is differentially calculated to obtain the altitude drop per unit time. The specific calculation process is: "Altitude drop per unit time" is equal to "altitude at the current moment" minus "altitude at the previous moment". Based on the altitude drop per unit time, the degree of air thinness is further evaluated. The degree of air thinness refers to the degree of decrease in the oxygen partial pressure in the air as the altitude increases. In a high-altitude environment, the greater the air thinness, the lower the oxygen partial pressure, and the higher the load on the cardiovascular system. The evaluation formula for the degree of air thinness is as follows: "Air pressure per unit time" is equal to "standard atmospheric pressure" multiplied by "1 minus altitude divided by 44330" to the power of "5.255". According to this formula, the air pressure changes at different altitudes can be calculated, and the air thinness data per unit time can be obtained by differential analysis, reflecting the thinness of oxygen supply in the air. The data of air thinness provides an important reference for the subsequent cardiovascular status assessment. The slowness of air pressure adaptation refers to the speed at which the human body adapts to the thin air (i.e., hypoxic environment). In a hypoxic environment, the human body needs a certain amount of time to adjust its physiological state to cope with the insufficient oxygen supply. In order to analyze the slowness of air pressure adaptation, the cumulative growth model was used to evaluate the patient's ability to adapt to air pressure at different altitudes in combination with the air thinness data. The specific approach is as follows: First, the air thinness data is combined with the patient's heart rate fluctuation data, and the slowness of air pressure adaptation is defined as the change in heart rate fluctuation per unit time, that is, "the slowness of air pressure adaptation" is equal to "the amplitude of change in heart rate fluctuation per unit time" divided by "the amplitude of change in air thinness per unit time". Based on the air thinness and air pressure adaptation slowness data, a short-term altitude emergency response intensity simulation is performed. This simulation aims to evaluate the emergency response intensity of the cardiovascular system when patients are exposed to a high altitude environment in a short period of time. By establishing a dynamic adjustment model, combined with short-term altitude changes and air pressure adaptation slowness, the heart rate response intensity of patients in a rapid altitude change environment is simulated. In the model, the patient's altitude change rate (unit: meters / minute) and air pressure adaptation slowness in a short period of time are taken into account to calculate the emergency response intensity. The key formula of the model is as follows: "Short-term altitude emergency response intensity" equals "air pressure adaptation slowness" multiplied by "altitude change amplitude" multiplied by "air thinness" plus "air pressure adaptation slowness" multiplied by "heart rate fluctuation amplitude per unit time". Through this formula, the short-term altitude emergency response intensity data can be obtained, so as to understand the patient's cardiovascular system's response ability under rapid altitude changes. Finally, combined with the short-term altitude emergency response intensity data and air pressure adaptation slowness data, the heart rate fluctuation data of stroke patients in exercise state are used to quantify the rapidity of breathing and heart rate.The quantification of rapid breathing and heart rate refers to the synchronous changes in breathing rate and heart rate in a high-altitude environment, which is manifested as the degree of rapid breathing. By combining the fluctuations of breathing rate and heart rate, the quantitative data of rapid breathing and heart rate are obtained by using the proportional analysis method. The formula is as follows: "Quantification of rapid breathing and heart rate" is equal to "the amplitude of heart rate change per unit time" divided by "the amplitude of respiratory rate change per unit time". Through this quantitative analysis, the comprehensive impact data of rapid breathing and heart rate changes in patients under different altitude environments can be obtained, which further reflects the patient's cardiovascular health status. Through this formula, the patient's ability to adapt to changes in air pressure can be quantitatively evaluated, and the slowness of air pressure adaptation data can be obtained, which provides a basis for further analysis of rapid breathing and heart rate.

[0043] Preferably, step S224 includes the following steps: Calculate the temperature and humidity variation coefficient for the temperature and humidity variation data of the patient's environment to obtain the temperature and humidity variation coefficient; The body temperature heat stress response is simulated based on the temperature and humidity variation coefficients of the heart rate fluctuation data of the stroke patients in motion state, and the body temperature heat stress response simulation data is obtained; Conducting thermoregulatory effect coupling on the body temperature heat stress response simulation data to obtain heat stress thermoregulatory effect data; According to the thermoregulatory effect data of heat stress, the thermoregulatory heart rate load increment data of the stroke patients in motion state were simulated to obtain the thermoregulatory heart rate load increment data.

[0044] In an embodiment of the present invention, first, the temperature and humidity data of the patient's environment are obtained, and these data are collected in real time by environmental monitoring sensors. The calculation of the environmental temperature and humidity change amplitude coefficient is to quantify the impact of temperature and humidity on body temperature regulation. By setting a fixed time step (for example, collecting temperature and humidity data once an hour), the temperature and humidity change amplitude within the time period is first calculated. The specific calculation formula is: "temperature and humidity change amplitude coefficient" is equal to "current moment temperature and humidity data" minus "previous moment temperature and humidity data" divided by "previous moment temperature and humidity data". Through this calculation method, the coefficient data of the temperature and humidity change amplitude can be obtained, reflecting the intensity of temperature and humidity changes, and providing basic data for subsequent body temperature heat stress response simulation. The body temperature heat stress response simulation is to combine the temperature and humidity change amplitude coefficient with the heart rate fluctuation data of stroke patients in motion state, and simulate the influence of the patient's body temperature regulation ability on heart rate changes under a specific temperature and humidity environment. The body temperature heat stress response simulation reflects the relationship between body temperature regulation and heart rate by establishing a physiological basic model. The model considers the impact of temperature and humidity changes on sweat glands, cardiovascular system and respiratory rate. The specific simulation process is as follows: "Body temperature heat stress response simulation" equals "temperature and humidity change amplitude coefficient" multiplied by "heart rate fluctuation amplitude under exercise" divided by "body temperature regulation speed". Through this simulation, the immediate impact of body temperature on heart rate can be quantified, thereby providing input for further simulation of body temperature regulation heart rate load increment. Body temperature regulation effect coupling refers to the interaction between the body temperature regulation mechanism (such as vascular contraction and expansion, sweat gland secretion, etc.) and heart rate fluctuation in a specific temperature and humidity environment. By analyzing the body temperature heat stress response simulation data and combining the temperature and humidity change amplitude coefficient, a coupling model of body temperature regulation effect and heart rate fluctuation is established. This coupling model considers that the human body regulates body temperature through blood flow and analyzes how body temperature changes affect the regulation effect of heart rate. The body temperature regulation effect coupling formula is: "heat stress body temperature regulation effect" equals "body temperature heat stress response simulation" multiplied by "body temperature regulation speed coefficient" minus "temperature and humidity change amplitude coefficient". Through this step, the data of the thermoregulation effect of heat stress can be obtained, which further reflects the linkage between thermoregulation and heart rate fluctuations. According to the data of the thermoregulation effect of heat stress, the incremental simulation of the thermoregulation heart rate load is performed. This simulation aims to evaluate the incremental effect of thermoregulation on heart rate load under a certain temperature and humidity environment. In this step, by combining the thermoregulation effect with the patient's exercise state heart rate fluctuation data, it simulates how thermoregulation affects the change of heart rate load in patients under exercise. The specific calculation formula is as follows: "Thermoregulation heart rate load increment" is equal to "heat stress thermoregulation effect" multiplied by "heart rate fluctuation amplitude under exercise state" divided by "coupling degree of thermoregulation and heart rate fluctuation". This simulation can quantify the incremental effect of thermoregulation on the patient's heart rate load through precise physiological modeling, providing a reliable basis for the final cardiovascular health assessment.Through the above steps, the impact of temperature and humidity changes on heart rate fluctuations of stroke patients during exercise can be comprehensively evaluated, thereby more accurately simulating the cardiovascular load of patients in different temperature and humidity environments, and providing detailed data support for stroke prevention.

[0045] Preferably, step S24 comprises the following steps: Step S241: performing triggering environment association with the triggering cause data of the stroke patient according to the cardiovascular environment influence data in the exercise state to obtain triggering cause environment association data; Step S242: performing abnormal blood dynamics simulation for stroke based on the cardiovascular environment influence data and the cause-triggered environment association data in the exercise state to obtain abnormal blood dynamics simulation data for stroke; Step S243: identifying environmental factor weighted risk factors for cardiovascular environmental impact data under exercise status according to the inducement triggering environment association data and stroke blood abnormality dynamic simulation data, and obtaining environmental factor weighted risk factors; Step S244: Calculate the probability of sudden stroke risk based on the triggering cause data of the stroke patient based on the weighted risk factor of the environmental factors, the abnormal blood dynamics simulation data of the stroke, and the triggering cause environment association data to obtain the sudden stroke risk probability data.

[0046] In an embodiment of the present invention, according to the cardiovascular environmental influence data of the patient in the state of motion, the triggering cause data of the stroke patient is subjected to association analysis to identify which external environmental factors have significant associations with the stroke causes. First, by analyzing the patient's heart rate fluctuations in the state of motion, temperature and humidity changes, air thinness and other environmental factors, combined with the records of stroke causes in historical case data, a set of association models between the triggering cause environment and the patient's physiological state is established. The model infers whether certain environmental conditions will trigger the cause of stroke by calculating the correlation between the patient's heart rate changes and environmental factors (for example, temperature and humidity, air pressure changes). The association model can use linear regression analysis or statistical correlation analysis to obtain "cause triggering environment association data" through data processing. After completing the analysis of the association of the cause triggering environment, the next step is to simulate the abnormal blood dynamics of stroke. The purpose of this step is to simulate the abnormal state of blood flow in stroke patients by combining the cardiovascular environmental influence data in the state of motion and the cause triggering environment association data. The abnormal blood dynamics simulation mainly focuses on the changes in physiological characteristics such as blood viscosity, blood flow velocity, and thrombosis. This simulation generates hemodynamic simulation data by considering the changes in blood flow under different environmental conditions (such as temperature, humidity, and altitude) and combining the individual characteristics of stroke patients. This process usually uses physiological modeling and hemodynamic models (such as the Navier-Stokes equation or fluid dynamics model) to simulate the changes in blood flow under stroke. Based on the abnormal blood dynamics simulation data of stroke, the next step is to identify the weighted risk factors of environmental factors for stroke. This process further analyzes the cause-trigger environmental association data and hemodynamic simulation results to identify which environmental factors have a greater impact on the triggering of stroke. Through regression analysis of multiple environmental factors (such as temperature and humidity changes, air pressure changes, etc.), the contribution of different environmental factors to stroke risk is calculated, and a weight coefficient is assigned to each environmental factor. This weight coefficient reflects the relative importance of each environmental factor to the risk of sudden stroke. For example, changes in temperature and humidity have a significant impact on blood pressure and heart rate fluctuations in some patients, while the thinness of the air has a greater impact on other patients. Through the identification of environmental factor weighted risk factors, the "environmental factor weighted risk factor" is finally obtained. The "weighted risk factor of environmental factors" is equal to the weighted average of the "regression coefficients of the impact of each environmental factor on the risk of stroke". This analysis step combines multiple regression analysis and weighted average method to quantify the impact of each environmental factor on the risk of stroke and obtain the most influential environmental factor weight data. Based on the "weighted risk factor of environmental factors" and "stroke blood abnormality dynamics simulation data" obtained from the above analysis, the probability of sudden stroke risk is finally calculated. The core of this calculation is to comprehensively evaluate the probability of stroke based on the weight of each environmental factor combined with the stroke blood abnormality dynamics simulation data.Through the risk calculation model of probability theory (such as logistic regression or Bayesian network), each environmental factor and its corresponding abnormal blood dynamics are weighted to obtain the risk probability of sudden stroke in patients. The final calculation result is the possibility of sudden stroke. Through this accurate risk probability prediction, timely warnings can be provided for stroke prevention. The "risk probability of sudden stroke" is equal to the weighted combination of "environmental factor weight risk factor" and "stroke blood abnormal dynamics simulation data".

[0047] Preferably, step S25 comprises the following steps: Step S251: quantifying the risk probabilities of the patient's environment data in different environments based on the stroke sudden risk probability data to obtain risk probability quantification data in different environments; Step S252: Calculating the mean difference of stress response degree of the patient's environment data according to the cardiovascular environment impact data in the exercise state to obtain the mean difference of environmental stress response degree; Step S253: performing scale normalization processing on the risk probability quantification data and the mean difference of environmental stress response degree under different environments to obtain risk quantification-environmental stress scale normalization data; Step S254: Calculate the local risk environment level of stroke for the patient's environmental data according to the risk quantification-environmental stress scale normalization data to obtain the local risk environment level of stroke.

[0048] In the embodiment of the present invention, the environmental data of the patient is further analyzed according to the "stroke sudden risk probability data" to quantify the stroke risk probability under different environmental conditions. In this process, the environmental data of the patient (including information such as temperature, humidity, air pressure, altitude, etc.) is first obtained, and compared with the stroke sudden risk probability data. In the analysis, the weighted average method is used to quantify the impact of various environmental factors on the stroke risk. For example, when the temperature rises or the air pressure changes drastically, the stroke sudden risk under each environmental condition is calculated in combination with the risk prediction results in the previous step. Finally, the risk probability quantification data under different environments is obtained. These data provide accurate environmental risk measurement for stroke prevention decisions. The degree of stress response of the patient's environmental data is calculated based on the "cardiovascular environmental impact data under exercise state". The mean difference in the degree of stress response reflects the difference in the patient's physiological response to changes in the external environment under different environmental conditions. In actual operation, the patient's heart rate fluctuation data in the exercise state is first combined with the environmental data to analyze the heart rate fluctuation and blood pressure changes under different environmental conditions. Then, by calculating the difference in the heart rate fluctuation amplitude of the patient under different environmental conditions, the mean difference in the degree of environmental stress response is obtained. The calculation formula is: mean difference of stress response = the difference between the standard deviation of heart rate fluctuation under each environmental condition and its mean. This process reflects the stress effect of environmental changes on the physiological state of stroke patients. The "risk probability quantification data under different environments" and the "mean difference of environmental stress response" are scaled and normalized. The purpose of scale normalization is to convert data of different dimensions into the same scale range for comparison and comprehensive analysis. Through normalization methods (such as minimum-maximum standardization method), the risk probability quantification data and stress response mean difference data under each environmental condition are normalized to the interval [0, 1]. This process ensures the comparability of data from different environments and avoids data distortion caused by different dimensions. Based on the "risk quantification-environmental stress scale normalization data", the local risk environment level of stroke in the patient's environment is calculated. The calculation of the local risk environment level of stroke is to combine the results of the above steps and the weights of each environmental factor to finally obtain the risk level of the environment in which the stroke patient is located. First, the normalized risk data and stress response data are weighted, and the environment is divided into multiple levels according to the weighted results. For example, the risk level can be divided into three levels: "low risk", "medium risk" and "high risk", and each level is divided according to the comprehensive evaluation of the above data. The specific level division is based on analyzing the stroke risk and the patient's stress response under various environmental conditions, and using the weighted average method to determine the corresponding environmental level.

[0049] Preferably, step S3 comprises the following steps: Step S31: sorting the patient's environment data in descending order of risk point distribution according to the stroke local risk environment level, to obtain descending order sorting data of the patient's environment risk points; Step S32: performing geographical environment stroke triggering trend analysis on the stroke sudden risk probability data according to the descending order sorting data of the environmental risk points, to obtain geographical environment stroke triggering trend data; Step S33: constructing a dynamic stroke prevention and early warning map based on the stroke triggering trend data of the geographical environment to obtain a dynamic stroke prevention and early warning map.

[0050] In the embodiment of the present invention, the environmental data of the patient is first sorted in descending order according to the "local risk environment level of stroke". Environmental risk points refer to specific environmental factor locations related to the probability of stroke, including areas with large temperature and humidity changes, areas with obvious altitude changes, etc. By sorting the environmental risk levels, high-risk areas can be identified, and data support can be provided for subsequent stroke risk analysis. The sorting process first matches all environmental data points (such as temperature, humidity, altitude, etc.) with the corresponding "local risk environment level of stroke". Then, these environmental data points are sorted from high to low according to the local risk environment level of stroke. The specific method used for sorting is to sort in reverse order based on environmental risk data to ensure that environmental points with higher risks are placed in front. Finally, the obtained "environmental risk point descending sorting data" can provide accurate risk point reference for the next step of geographical environment analysis. By descending sorting, the risk level corresponding to each environmental point can be determined, thereby identifying the most dangerous environmental area, and then providing a basis for risk warning. Using the "environmental risk point descending sorting data", the "stroke sudden risk probability data" is analyzed for geographical environment to obtain stroke triggering trend data. The goal of this analysis is to evaluate the potential impact of environmental changes in different geographical locations on the risk of sudden stroke. The analysis process first obtains sorted environmental risk point data and combines it with stroke sudden risk probability data to identify risk trends in different geographical areas. The specific method is: based on the sorted environmental data points, trend analysis algorithms (such as polynomial regression analysis or weighted moving average method) are applied to model the risk probability of each environmental point in different time periods to identify the risk trends of different environmental points. This process can reveal which areas have a higher probability of stroke in a certain period of time and help identify high-risk areas for stroke. The analysis results will generate "geographic environment stroke triggering trend data", which reflects the potential trend and time distribution of stroke sudden events in different geographical environments, providing support for the subsequent construction of dynamic early warning maps. Based on the "geographic environment stroke triggering trend data", a "dynamic stroke prevention and early warning map" is constructed. The core function of this map is to provide stroke risk assessment in different geographical areas through dynamic updates and real-time data input, and to monitor the impact of environmental changes on stroke risk in real time.

[0051] The construction process first needs to combine the "geographical environment stroke trigger trend data" with real-time environmental data, and use data visualization technology (such as heat maps or regional division maps) to visualize the stroke risk in each region. Heat maps can distinguish the high and low stroke risks by different colors, marking high-risk areas as red and low-risk areas as green, so that the risk levels of different regions are clear at a glance. At the same time, the dynamic map function is used to update the stroke warning map in a timely manner according to the real-time environmental data changes. For example, when the temperature and humidity in a certain area change significantly, the stroke risk in the area will be automatically updated and displayed on the map. The dynamic update algorithm is based on the input of real-time data, and the map content is updated regularly to ensure that the map reflects the latest stroke risk information. Through this step, a dynamic and real-time stroke prevention and warning map can be constructed based on the stroke risk trends in different regions, helping relevant personnel to understand the risk situation in each region in a timely manner and take corresponding preventive measures.

[0052] Preferably, step S32 includes the following steps: Step S321: performing gradient risk increment analysis between adjacent risk points on the descending sorted data of environmental risk points and the stroke sudden risk probability data based on a recurrent neural network in artificial intelligence to obtain gradient risk increment data between adjacent risk points; Step S322: performing potential correlation analysis on the gradient risk increment data between adjacent risk points to obtain risk potential correlation data between adjacent risk points; Step S323: Performing geographical environment stroke triggering trend analysis based on the risk potential correlation data between adjacent risk points to obtain geographical environment stroke triggering trend data.

[0053] In an embodiment of the present invention, a recurrent neural network (RNN) in artificial intelligence is used to perform a gradient risk increment analysis between adjacent risk points on the "environmental risk point descending sorting data" and the "stroke sudden risk probability data". The gradient risk increment refers to the rate of change of the stroke sudden risk probability in the adjacent area of ​​the environmental risk point. First, a recurrent neural network model is used for training. The input of the model is the environmental data points and the corresponding risk probabilities in the "environmental risk point descending sorting data" and the "stroke sudden risk probability data". RNN is suitable for processing time series data. Therefore, in this step, the environmental risk data is converted into a time series input, and the gradient change between adjacent risk points is considered. The model analyzes the adjacent data through a feedback mechanism. The specific gradient risk increment analysis process is: through the recurrent neural network model, the gradient calculation is performed on the data of each pair of adjacent environmental risk points, that is, the difference in the stroke sudden risk probability between adjacent points. The risk increment value between each adjacent risk point is obtained by using the regression ability of the recurrent neural network. These gradient risk increment data can reflect the sensitivity of adjacent environmental points to the change of stroke risk probability. In this step, the potential correlation analysis of the "gradient risk increment data between adjacent risk points" is performed to identify the correlation between different environmental risk points. Specifically, the Pearson correlation coefficient or Spearman rank correlation coefficient is used to measure the correlation between adjacent risk points. First, the correlation coefficient between the risk increments of adjacent environmental points is calculated by statistically analyzing the gradient risk increment data. This analysis aims to identify those environmental points where the risk of sudden stroke shows similar trends under environmental changes. The operation process of the potential correlation analysis is: extract the incremental value pairs of adjacent risk points from the gradient risk increment data and calculate the correlation between them. If the incremental data of two points have the same changing trend in the same time period, it is considered that there is a strong potential correlation between them. In this way, the "risk potential correlation data between adjacent risk points" can be obtained, which provides strong support for the subsequent triggering trend analysis. Based on the "risk potential correlation data between adjacent risk points", the geographical environment stroke triggering trend analysis is carried out to reveal the trend of sudden stroke risk in different environments over time. Specifically, the risk potential correlation data is modeled using a time series-based trend analysis method, such as the weighted moving average method or the exponential smoothing method. First, based on the potential correlation data identified in the previous step, an analysis model is constructed to determine which geographical environmental factors (such as temperature, humidity, altitude, etc.) are closely related to the risk of sudden stroke. Then, the changes in these factors are simulated through trend analysis to evaluate the upward or downward trend of stroke risk. The goal of this analysis process is to identify the triggering trend of sudden stroke risk in different environments.Through trend analysis, we can obtain geographical environment stroke triggering trend data, which can accurately reflect the changing pattern of sudden stroke risk in a specific environment and provide reference data for the construction of further dynamic early warning maps.

[0054] Therefore, the embodiments should be regarded as illustrative and non-restrictive from all points, and the scope of the present invention is limited by the appended claims rather than the above description, and it is therefore intended that all changes falling within the meaning and range of equivalent elements of the application documents are included in the present invention.

[0055] The above description is only a specific embodiment of the present invention, so that those skilled in the art can understand or implement the present invention. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but should conform to the widest scope consistent with the principles and novel features invented herein.

Claims

1. A method for constructing a stroke prevention map system based on artificial intelligence, characterized in that: The following steps are involved: Step S1: collecting real-time physiological sign data of stroke patients through wearable devices to obtain real-time sign filling data of stroke patients; collecting historical case data of stroke patients to obtain desensitized historical case data of stroke patients; Step S2: extracting the patient's environment from the real-time vital sign filling data of the stroke patient to obtain the patient's environment data; performing case triggering cause analysis on the desensitized data of the historical cases of the stroke patient to obtain the triggering cause data of the stroke patient; The stroke sudden risk probability is calculated based on the triggering cause data of the stroke patient to obtain the stroke sudden risk probability data; based on the stroke sudden risk probability data, the stroke local risk environment level is calculated based on the patient's environmental data to obtain the stroke local risk environment level; Step S3: performing geographical environment stroke triggering trend analysis on the stroke sudden risk probability data according to the stroke local risk environment level to obtain geographical environment stroke triggering trend data; Based on the stroke triggering trend data of the geographical environment, a dynamic stroke prevention and early warning map is constructed to obtain a dynamic stroke prevention and early warning map; Step S4: respond to the risk assessment results of the dynamic stroke prevention and early warning map based on the stroke triggering trend data of the geographical environment to obtain risk assessment result response data, and use the preset medical resource matching model to push the risk assessment result response data to the referral agency to obtain stroke risk referral agency push data.

2. The method for constructing a stroke prevention map system based on artificial intelligence according to claim 1, characterized in that: Step S1 includes the following steps: Step S11: collecting real-time physiological sign data of stroke patients through wearable devices to obtain real-time physiological sign data of stroke patients; Step S12: collecting historical case data of stroke patients to obtain historical case data of stroke patients; Step S13: desensitizing the real-time physiological sign data of the stroke patients and the historical case data of the stroke patients, respectively, to obtain desensitized real-time physiological sign data of the stroke patients and desensitized historical case data of the stroke patients; Step S14: Fill missing values ​​in the desensitized real-time vital sign data of stroke patients to obtain real-time vital sign filled data of stroke patients.

3. The method for constructing a stroke prevention map system based on artificial intelligence according to claim 1, characterized in that: Step S2 includes the following steps: Step S21: extracting the patient's environment from the real-time vital sign filling data of the stroke patient to obtain the patient's environment data; Step S22: performing cardiovascular environmental impact assessment on the real-time vital sign filled data of the stroke patient in the exercise state according to the patient's environment data, to obtain cardiovascular environmental impact data in the exercise state; Step S23: performing case triggering cause analysis on the desensitized historical case data of stroke patients to obtain triggering cause data of stroke patients; Step S24: calculating the probability of sudden stroke risk for the stroke patient's triggering cause data according to the cardiovascular environment influence data in the exercise state, and obtaining the sudden stroke risk probability data; Step S25: Calculate the local risk environment level of stroke for the patient's environmental data based on the stroke sudden risk probability data and the cardiovascular environmental impact data under exercise status to obtain the local risk environment level of stroke.

4. The method for constructing a stroke prevention map system based on artificial intelligence according to claim 3, characterized in that: Step S22 includes the following steps: Step S221: analyzing the heart rate fluctuation of the stroke patient in the exercise state on the real-time vital sign filling data to obtain the heart rate fluctuation data of the stroke patient in the exercise state; Step S222: extracting the temperature, humidity and altitude change data of the patient's environment to obtain the temperature and humidity change data of the patient's environment and the altitude change data of the patient; Step S223: quantifying the rapid breathing and heart rate of the stroke patient's exercise state heart rate fluctuation data according to the patient's altitude change data to obtain the rapid breathing and heart rate quantification data; Step S224: performing a temperature regulation heart rate load increment simulation on the heart rate fluctuation data of the stroke patient in motion based on the temperature and humidity change data of the patient's environment to obtain the temperature regulation heart rate load increment data; Step S225: Evaluate the cardiovascular environmental impact under exercise conditions based on the rapid breathing and heart rate quantification data and the temperature regulation heart rate load increment data to obtain cardiovascular environmental impact data under exercise conditions.

5. The method for constructing a stroke prevention map system based on artificial intelligence according to claim 4, characterized in that: Step S223 includes the following steps: The altitude difference per unit time is calculated for the patient's altitude change data using a preset time step to obtain the altitude difference per unit time; Evaluate the air thinness degree of the altitude difference per unit time to obtain the air thinness degree data; According to the data of air thinness, the slowness of air pressure adaptation is analyzed on the heart rate fluctuation data of stroke patients in motion state, and the slowness of air pressure adaptation of heart rate state is obtained; Based on the data of air thinness and the slowness of heart rate adaptation to air pressure, the short-term altitude emergency response intensity was simulated on the heart rate fluctuation data of stroke patients in motion, and the short-term altitude emergency response intensity data was obtained. According to the short-term altitude emergency reaction intensity data and the heart rate state air pressure adaptation slowness data, the heart rate fluctuation data of the stroke patients in the exercise state are used to quantify the rapid breathing and heart rate, and the rapid breathing and heart rate quantitative data are obtained.

6. The method for constructing a stroke prevention map system based on artificial intelligence according to claim 4, characterized in that: Step S224 includes the following steps: Calculate the temperature and humidity variation coefficient for the temperature and humidity variation data of the patient's environment to obtain the temperature and humidity variation coefficient; The body temperature heat stress response is simulated based on the temperature and humidity variation coefficients of the heart rate fluctuation data of the stroke patients in motion state, and the body temperature heat stress response simulation data is obtained; Conducting thermoregulatory effect coupling on the body temperature heat stress response simulation data to obtain heat stress thermoregulatory effect data; According to the thermoregulatory effect data of heat stress, the thermoregulatory heart rate load increment data of the stroke patients in motion state were simulated to obtain the thermoregulatory heart rate load increment data.

7. The method for constructing a stroke prevention map system based on artificial intelligence according to claim 3, characterized in that: Step S24 includes the following steps: Step S241: performing triggering environment association with the triggering cause data of the stroke patient according to the cardiovascular environment influence data in the exercise state to obtain triggering cause environment association data; Step S242: performing abnormal blood dynamics simulation for stroke based on the cardiovascular environment influence data and the cause-triggered environment association data in the exercise state to obtain abnormal blood dynamics simulation data for stroke; Step S243: performing environmental factor weighted risk factor identification on the cardiovascular environmental impact data under exercise state according to the inducement triggering environment association data and the stroke blood abnormality dynamic simulation data to obtain the environmental factor weighted risk factor; Step S244: Calculate the probability of sudden stroke risk based on the triggering cause data of the stroke patient based on the weighted risk factor of the environmental factors, the abnormal blood dynamics simulation data of the stroke, and the triggering cause environment association data to obtain the sudden stroke risk probability data.

8. The method for constructing a stroke prevention map system based on artificial intelligence according to claim 3, characterized in that: Step S25 includes the following steps: Step S251: quantifying the risk probabilities of the patient's environment data in different environments based on the stroke sudden risk probability data to obtain risk probability quantification data in different environments; Step S252: Calculating the mean difference of stress response degree of the patient's environment data according to the cardiovascular environment impact data in the exercise state to obtain the mean difference of environmental stress response degree; Step S253: performing scale normalization processing on the risk probability quantification data and the mean difference of environmental stress response degree under different environments to obtain risk quantification-environmental stress scale normalization data; Step S254: Calculate the local risk environment level of stroke for the patient's environmental data according to the risk quantification-environmental stress scale normalization data to obtain the local risk environment level of stroke.

9. The method for constructing a stroke prevention map system based on artificial intelligence according to claim 1, characterized in that: Step S3 includes the following steps: Step S31: sorting the patient's environment data in descending order of risk point distribution according to the stroke local risk environment level, to obtain descending order sorting data of the patient's environment risk points; Step S32: performing geographical environment stroke triggering trend analysis on the stroke sudden risk probability data according to the descending order sorting data of the environmental risk points, to obtain geographical environment stroke triggering trend data; Step S33: constructing a dynamic stroke prevention and early warning map based on the stroke triggering trend data of the geographical environment to obtain a dynamic stroke prevention and early warning map.

10. The method for constructing a stroke prevention map system based on artificial intelligence according to claim 8, characterized in that: Step S32 includes the following steps: Step S321: performing gradient risk increment analysis between adjacent risk points on the descending sorted data of environmental risk points and the stroke sudden risk probability data based on a recurrent neural network in artificial intelligence to obtain gradient risk increment data between adjacent risk points; Step S322: performing potential correlation analysis on the gradient risk increment data between adjacent risk points to obtain risk potential correlation data between adjacent risk points; Step S323: Performing geographical environment stroke triggering trend analysis based on the risk potential correlation data between adjacent risk points to obtain geographical environment stroke triggering trend data.

Images