Blood pressure management system, method and equipment based on dynamic environment perception

Abstract

The invention discloses a blood pressure management system based on dynamic environment perception, and the system comprises an environment perception module which is used for monitoring environment parameters including temperature, humidity, air pressure and noise in real time, and is in linkage with an external environment adjustment system through multi-parameter coupling compensation to optimize and adjust a measurement environment to a specified state; the data processing and control module is used for carrying out deep processing and analysis on the collected data and making a control decision according to an analysis result so as to realize accurate management and monitoring on the blood pressure; and the blood pressure measurement and interaction module is used for executing blood pressure measurement operation and feeding back a measurement result to the user. By integrating the environment sensing technology, the measurement parameters are adjusted in real time to adapt to different environment conditions, so that the accuracy of blood pressure measurement is improved, the device can sense and adapt to various environment changes, and it is ensured that the blood pressure can be accurately monitored in various environments.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent blood pressure management, and particularly to a blood pressure management system, method and device based on dynamic environment perception. Background Art

[0002] Intelligent blood pressure management is an interdisciplinary field integrating medicine, electronic engineering and artificial intelligence, focusing on improving the convenience, accuracy and efficiency of blood pressure monitoring through technological innovation. With the aging of the population and the rising incidence of chronic diseases, intelligent blood pressure management technology has received increasing attention globally and has become a key means of preventing cardiovascular diseases and improving public health levels.

[0003] The latest technologies in the current intelligent blood pressure management field mainly include:

[0004] (1) Ambulatory blood pressure monitoring technology: realizing 24-hour continuous blood pressure monitoring through wearable devices, capturing blood pressure fluctuations more accurately, especially helpful for detecting masked hypertension and nocturnal hypertension;

[0005] (2) Application of artificial intelligence: using AI technology to improve the accuracy of blood pressure measurement, such as more accurately measuring blood pressure by analyzing pulse sounds, and maintaining high accuracy even in complex situations;

[0006] (3) New wearable blood pressure measurement devices: combining fashion and function, such as blood pressure monitoring devices in the form of smart watches, which are convenient for users to wear daily and realize non-invasive monitoring;

[0007] (4) Remote monitoring and data management: combining Internet and big data technologies to realize remote transmission and intelligent management of blood pressure data, facilitating real-time monitoring and adjustment of treatment plans by doctors and patients.

[0008] Although the existing intelligent blood pressure management technology has made certain progress, there are still defects. Many existing wearable blood pressure monitoring devices still have deficiencies in measurement accuracy. Especially during dynamic monitoring or environmental changes, they tend to ignore the potential impact of environmental factors on blood pressure readings, and the measurement results may be interfered, resulting in possible deviations in the measurement results. Moreover, the existing blood pressure monitoring devices often cannot well adapt to different environmental conditions, such as temperature changes, humidity, noise, etc. These factors may affect the accuracy of the measurement results. In addition, traditional blood pressure monitors lack personalized measurement schemes and cannot adjust measurement parameters according to the specific situation of users, and usually do not have the ability to monitor and analyze data in real time.

[0009] Therefore, this application specifically proposes a blood pressure management system based on dynamic environment perception to solve the above technical problems. Summary of the Invention

[0010] The main purpose of the present invention is to provide a blood pressure management system based on dynamic environmental perception. By integrating environmental perception technology, the measurement parameters are adjusted in real time to adapt to different environmental conditions, thereby improving the accuracy of blood pressure measurement, and enabling the equipment to perceive and adapt to various environmental changes, ensuring accurate blood pressure monitoring in various environments, so as to solve the technical problems raised in the background technology.

[0011] The present invention adopts the following technical solutions to solve the above technical problems:

[0012] A blood pressure management system based on dynamic environment perception, comprising:

[0013] The environmental perception module is used to monitor environmental parameters including temperature, humidity, air pressure, and noise in real time, and optimize the measurement environment to a specified state through multi-parameter coupling compensation and linkage with the external environmental adjustment system;

[0014] Data processing and control module, used to deeply process and analyze the collected data, and make control decisions based on the analysis results to achieve accurate management and monitoring of blood pressure;

[0015] The blood pressure measurement and interaction module is used to perform blood pressure measurement operations and feed back the measurement results to the data processing and control module and the user.

[0016] Preferably, the control logic of the multi-parameter coupling compensation linkage control of the external environment of the dynamic environment perception module includes:

[0017] When the temperature is greater than 28°C and the humidity is greater than 80%, a comprehensive compensation strategy of vasodilation compensation factor γ1×blood viscosity correction coefficient γ2 is used to compensate the blood pressure measurement result δ0, and the blood pressure measurement result δ l is: l =δ0×γ1×γ2;

[0018] When the noise level is greater than 70dB and the light intensity is greater than 1500lux, the dual-channel filtering algorithm is activated to eliminate artifact signals.

[0019] Build an environmental compensation system to generate the environmental interference index EEI in real time based on the ISO 81060-3 standard certification. When EEI>0.85, the environmental adjustment command is automatically triggered, and the external environmental adjustment system is linked to measure the environment to the specified state.

[0020] Preferably, the data processing and control module is internally provided with:

[0021] A data receiving and preprocessing unit, used to receive real-time data from the environment sensing module and the blood pressure measurement and interaction module, and perform data preprocessing operations;

[0022] A data processing and control unit for multi-source data fusion and correction of environmental parameters and physiological data, dynamically perceiving environmental parameters and user physiological signals, dynamically correcting the original blood pressure value to eliminate environmental interference and output the optimized blood pressure value BP adj , and constructing an adaptive algorithm framework based on Kalman filtering to achieve noise suppression through a state space model;

[0023] A data analysis and prediction unit for analyzing and extracting specified information from environmental data and blood pressure data, and performing deep learning in combination with historical data to predict blood pressure changes.

[0024] Preferably, the specific operation process of the data processing and control unit for multi-source data fusion and correction includes:

[0025] a1. Generating an environmental feature matrix E within a time period t by dynamically perceiving environmental parameters including temperature, humidity, air pressure, noise, and light, where: t , we have:

[0026] E t =[T t ,H t ,P t ,N t ,L t

[0027] where T t is the temperature parameter, H t is the humidity parameter, P t is the air pressure parameter, N t is the noise parameter, L t is the light parameter;

[0028] a2. Conducting environmental impact factor modeling and training the weight coefficient w using multiple linear regression or random forest algorithm based on the historical data set to minimize the correction error, where: i , we have:

[0029] min∑(BP true -(BP raw -∑w i ·ΔE i )) 2

[0030] ΔE i =E i,current -E i,baseline

[0031] where BP true is the regularly collected blood pressure calibration value, BP raw is the originally collected blood pressure value, and ΔE i ​is the error between the actual and ideal environmental parameters, E i,current is the actual environmental parameter, E i,baseline is the ideal environmental parameter;

[0032] a3. During blood pressure calibration and verification, dynamically adjust the calibration amplitude k according to the individual age differences of the users whose blood pressure is measured to obtain the optimized blood pressure value BP adj , there is:

[0033]

[0034] Among them, age represents age, and β is the attenuation factor, which is obtained by fitting the historical data of the users whose blood pressure is measured.

[0035] Preferably, the specific operation process of using multiple linear regression to train the weight coefficient in step a2 includes: setting a random small value as the initial regression coefficient, and using the gradient descent method to iterate and optimize. By calculating the gradient of the loss function with respect to the regression coefficient and updating the coefficient in the opposite direction of the gradient until the loss function converges to the minimum value, so as to determine the optimal weight coefficient w i .

[0036] Preferably, the specific operation process of using the random forest algorithm to train the weight coefficient in step a2 includes: determining the number of decision trees according to the dataset size and computing resources, then constructing and training multiple decision trees through sampling with replacement of the samples, and finally obtaining the optimal weight coefficient w of the final prediction from the results of the decision trees i .

[0037] Preferably, the specific operation process of the data analysis and prediction unit for deep learning includes:

[0038] b1. Based on the time-series environmental data E t , the time-series corrected blood pressure value BP adj,t and the profile data containing the user's health information, construct an LSTM prediction model;

[0039] b2. The input layer of the LSTM prediction model receives the time-series data window [E t-n , BP adj,t-n ,…, E t , BP adj,t , and performs sliding window sampling, where the window size n is set to 6, which is used to represent the data of the past 1 hour;

[0040] b3. Set two stacked LSTM layers in the LSTM prediction model, with 64 neurons in each layer, and add Dropout = 0.2 to prevent overfitting;

[0041] b4. Output the blood pressure prediction value BP at the future time point t + 1 through the fully connected layerpred and optimizing the environmental regulation strategy through an auxiliary fitness function.

[0042] Preferably, the specific training process of the LSTM prediction model includes:

[0043] Performing data augmentation on historical data by means of sliding window sampling and adding Gaussian noise σ = 1 mmHg to improve the generalization ability of the model. At this time, the loss function Γ is specifically:

[0044]

[0045] where BP pred,i is the predicted blood pressure value at the i-th time, BP ture,i is the calibrated blood pressure value collected regularly at the i-th time, and λ is the intensity adjustment value of L2 regularization;

[0046] For the user's health record, freeze the underlying parameters of the LSTM and only train the top fully connected layer to obtain the auxiliary function:

[0047]

[0048] where BP adj is the optimized blood pressure value, BP true is the calibrated blood pressure value collected regularly, ΔE i is the error between the actual and ideal environmental parameters, and BP target is the target blood pressure range set by the user;

[0049] In the auxiliary function, |BP adj - BP ture |, ∑ i |ΔE i |, |BP pred - BP target | represent measurement error, environmental regulation cost, and trend deviation respectively. λ1, λ2, and λ3 represent adjustable influence factors of measurement error, environmental regulation cost, and trend deviation respectively, and λ1 + λ2 + λ3 = 1.

[0050] On the other hand, the present invention also discloses a blood pressure management method, which uses any one of the above-mentioned blood pressure management systems based on dynamic environment perception to perform operations, including:

[0051] Real-time monitoring of environmental information and regulating the environment through the environmental perception module, and then transmitting the data to the data processing and control module;

[0052] The data processing and control module adjusts the blood pressure measurement strategy according to the environmental data, and then sends instructions to the blood pressure measurement and interaction module;

[0053] The blood pressure measurement and interaction module controls the execution of blood pressure measurement operations and feeds back the results to the user.

[0054] In another aspect, the present invention further discloses a blood pressure management device, characterized in that it includes an interactive interface, a processor, a sphygmomanometer body, a cuff, and a measurement sensor, wherein:

[0055] The interactive interface, the main body of the sphygmomanometer, and the measuring sensor are all connected to the processor;

[0056] The interactive interface and the cuff are both arranged on the sphygmomanometer body, and the sphygmomanometer body and the measuring sensor are used to contact the user to perform blood pressure measurement operations.

[0057] The processor includes a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, the processor performs the method steps of blood pressure management and detection based on any of the above-mentioned systems.

[0058] It can be seen from the above technical solution that the present invention provides a blood pressure management system based on dynamic environment perception. Compared with the prior art, the present invention has the following advantages:

[0059] 1. The present invention can monitor environmental parameters such as temperature, humidity, air pressure, noise, etc. in real time and link the external environment adjustment system to actively optimize the measurement environment, thereby eliminating the negative impact of environmental interference factors on blood pressure measurement and ultimately improving the accuracy of blood pressure measurement.

[0060] 2. The present invention can conduct multi-dimensional analysis based on environmental parameters and user physiological signals, and accurately correct the original blood pressure values through multiple linear regression, random forest algorithm and dynamic adjustment strategy based on age differences, thereby improving the reliability and personalized adaptability of blood pressure data, and ultimately achieving precise and scientific blood pressure management.

[0061] 3. The present invention can predict the trend of blood pressure changes and balance measurement errors, environmental adjustment costs and trend deviations, thereby improving the resource utilization efficiency of health management and realizing active blood pressure risk prevention and control.

[0062] 4. The present invention integrates an interactive interface, processor and sensor into the blood pressure management device, and is equipped with an adaptive algorithm and environmental adjustment linkage mechanism, so as to provide real-time, accurate and personalized blood pressure monitoring and environmental optimization functions, thereby enhancing the convenience and effectiveness of user health management, and ultimately significantly improving the health and quality of life of patients with hypertension.

[0063] It should be understood that the content described in this part is not intended to identify the key or important features of the embodiments of the present invention, nor is it used to limit the scope of the present invention. Other features of the present invention will become easily understood through the following description. Of course, any product implementing the present invention does not necessarily need to achieve all the above-mentioned advantages simultaneously. BRIEF DESCRIPTION OF THE DRAWINGS

[0064] The accompanying drawings forming a part of this application are used to provide a further understanding of the present invention. The schematic embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation to the present invention. In the drawings:

[0065] Figure 1 is the overall system structure block diagram of the present invention;

[0066] Figure 2 is the internal structure block diagram of the data processing and control module of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0067] The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Without conflict, the embodiments in this application and the features in the embodiments can be combined with each other. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0068] In the embodiments, please refer in detail to Figures 1 to 2 .

[0069] As Figure 1 shown, a blood pressure management system based on dynamic environment perception proposed in the embodiments of the present invention includes an environment perception module, a data processing and control module, and a blood pressure measurement and interaction module. Each module cooperates with each other to form a closed-loop control system. Specifically:

[0070] (1) The environment perception module is used to monitor environmental parameters (temperature, humidity, air pressure, noise, etc.) in real time, provide environmental data support for blood pressure measurement, and through multi-parameter coupling compensation, link the external environment adjustment system to optimize and adjust the measurement environment to a specified state.

[0071] At this time, the dynamic environment perception module serves as the environmental data center of the blood pressure management system. Its core lies in determining the complex influence mechanism of environmental physiology on blood pressure measurement through multi-dimensional environmental parameter monitoring, and each environmental parameter has different action paths and quantitative influences on blood pressure measurement.

[0072] Among them, the dynamic environment perception module significantly improves the accuracy of blood pressure measurement through precise monitoring range and advanced algorithm compensation:

[0073] In terms of temperature parameters, the monitoring range is 10 - 40°C ± 0.5°C. The ambient temperature directly affects the peripheral vascular tone through the thermal regulation system;

[0074] When the temperature > 28°C, the skin blood vessel dilation rate increases by 30%, which will cause the redistribution of blood volume, resulting in a 5 - 10 mmHg decrease in systolic blood pressure and a 3 - 5 mmHg decrease in diastolic blood pressure; when the temperature < 18°C, the sympathetic nerve excitement causes the catecholamine secretion to increase by 40%, which in turn leads to a falsely high blood pressure reading. The measurement error of an electronic blood pressure monitor without temperature compensation in a 32°C environment reaches ±8 mmHg, while through the temperature compensation algorithm of this module, the error can be effectively controlled within ±2 mmHg. The monitoring range of humidity parameters is 20 - 90% RH ± 3%.

[0075] When the relative humidity > 70%, the high - humidity environment will increase the blood viscosity by 8%, which in turn indirectly affects the calculation of pulse wave velocity. For every 10% increase in humidity, the standard deviation of the ambulatory blood pressure monitor expands by 0.7 mmHg. During the plum rain season (humidity 85 - 95%), a special filtering algorithm needs to be activated to maintain the measurement accuracy.

[0076] The monitoring range of barometric pressure parameters is 300 - 1100 hPa ± 1.5%; for every 100 - meter increase in altitude (the barometric pressure decreases by 12 hPa), the blood oxygen saturation drops by 1.2%, triggering a compensatory increase in heart rate of 5 - 10 bpm; this module can automatically correct altitude - related errors through the built - in barometric pressure - altitude conversion model. In a 3000 - meter plateau environment, the accuracy of blood pressure measurement can be improved by 23%.

[0077] The monitoring range of noise parameters is 30 - 90 dB ± 2 dB. When the ambient noise is above 65 dB, the cortisol level increases by 35%, resulting in an instantaneous systolic blood pressure fluctuation of 10 - 15 mmHg. This module adopts adaptive noise reduction technology, and even in an 80 - dB subway environment, it can still ensure the measurement stability of ±1.5 mmHg.

[0078] The monitoring range of light parameters is 0 - 2000 lux ± 5%. Strong light (> 1000 lux) affects the circadian rhythm through the suprachiasmatic nucleus, which will reduce the nocturnal blood pressure drop by 8 mmHg. The module is equipped with an intelligent light adaptation system, which can dynamically adjust the measurement time window according to the light intensity to ensure the rhythm accuracy of 24 - hour ambulatory blood pressure monitoring.

[0079] In addition, this module innovatively constructs a multi-parameter coupling compensation algorithm: when the temperature > 28°C and the humidity > 80%, a comprehensive compensation strategy of the vasodilation compensation factor (γ1 = 1.15) × the blood viscosity correction coefficient (γ2 = 0.93) is adopted to compensate the blood pressure measurement result δ0, and the blood pressure measurement result δ is obtained. l It is: δ l = δ0 × γ1 × γ2;

[0080] In the state where the noise > 70 dB and the illumination > 1500 lux, the dual-channel filtering algorithm is activated, which can eliminate 12 - 15% of the artifact signals. In specific actual use, after being verified by clinical applications, after enabling the environmental perception compensation, the standard deviation of daytime blood pressure measurement drops from 8.7 mmHg to 3.2 mmHg, the detection accuracy of the early morning blood pressure peak increases by 42%, and the misjudgment rate of orthostatic hypotension decreases by 68%.

[0081] An environmental compensation system is constructed. Based on the ISO81060-3 standard certification, the environmental interference index EEI is generated in real time. When EEI > 0.85, the environmental adjustment instruction is automatically triggered, and the external environmental adjustment system is linked to adjust the measurement environment to the specified ideal state (temperature 22 ± 1°C, humidity 50 ± 5%, noise < 40 dB).

[0082] This active environmental control enables blood pressure measurement to reach a high level of accuracy, providing reliable environmental benchmark data for personalized hypertension management.

[0083] In addition, it should be noted that in this way, the environmental perception module provides comprehensive environmental information for the blood pressure management system, making blood pressure measurement more accurate, and at the same time providing a more comfortable and suitable measurement environment for users.

[0084] The environmental perception module is the front end of the entire system. It provides real-time environmental data for the data processing and control module to ensure the accuracy of blood pressure measurement results.

[0085] At the same time, by setting a dynamic environmental perception module in the system, environmental parameters such as temperature, humidity, air pressure, and noise can be monitored in real time and the external environmental adjustment system can be linked to actively optimize the measurement environment, thereby eliminating the negative impact of environmental interference factors on blood pressure measurement and ultimately improving the accuracy of blood pressure measurement.

[0086] (2) The data processing and control module is used to deeply process and analyze the collected data, and make control decisions based on the analysis results to achieve precise management and monitoring of blood pressure.

[0087] Specifically, referring to Figure 2 , the work of this module can be divided into several key functional sub-units, specifically including:

[0088] (a) A data reception and processing unit for receiving real-time data from an environmental perception module and a blood pressure measurement, including sensor data such as environmental temperature, humidity, noise, air quality, light intensity, etc., and the user's blood pressure data, which are usually transmitted wirelessly to the processing module, possibly through standard communication protocols such as Bluetooth and Wi-Fi.

[0089] The first step after data reception at this time is preprocessing, mainly including operations such as noise removal, data verification, and format conversion to ensure the accuracy and consistency of the data.

[0090] (b) A data processing and control unit for performing multi-source data fusion and correction of environmental parameters and physiological data. By dynamically perceiving environmental parameters and the user's physiological signals, it dynamically corrects the original blood pressure value to eliminate environmental interference and outputs an optimized blood pressure value BP adj , and constructs an adaptive algorithm framework based on Kalman filtering to achieve noise suppression through a state space model.

[0091] Among them, at the data input end, the system uses a programmable analog filter to perform physical layer preprocessing on the original signal: the temperature and humidity sensor channels are configured with an eighth-order low-pass filter with a cut-off frequency of 0.1 Hz to effectively filter out power frequency interference; the noise sensor adopts a 20 Hz - 20 kHz band-pass filtering scheme to accurately match the auditory characteristics of the human ear; the optical sensor deploys an adaptive notch filter to dynamically eliminate 50 / 60 Hz fluorescent lamp stroboscopic noise.

[0092] At the same time, for the time-varying characteristics of dynamic environmental parameters, the system constructs an adaptive algorithm framework based on Kalman filtering to achieve noise suppression through a state space model.

[0093] Furthermore, the specific operation process of the data processing and control unit for multi-source data fusion and correction includes:

[0094] a1. By dynamically perceiving environmental parameters including temperature, humidity, air pressure, noise, and light, these data are collected by distributed temperature and humidity sensors, barometers, noise sensors, and light sensors, while the user's physiological signals include body surface temperature and micro-expressions. The body surface temperature distribution is captured and analyzed by an infrared thermal imaging camera. After data collection, preprocessing work will be carried out: first, the Kalman filtering method is used to eliminate sensor noise to achieve filtering and noise reduction, and then normalization processing is performed to map each parameter to a standard range, such as a temperature range of 0 - 1 corresponding to 10℃ - 40℃;

[0095] After these processes, an environmental feature matrix E within the time period t is generated t , and there is:

[0096] E t =[Tt , H t , P t , N t , L t

[0097] Among them, T t is the temperature parameter, H t is the humidity parameter, P t is the air pressure parameter, N t is the noise parameter, L t is the light intensity parameter;

[0098] a2. Perform environmental impact factor modeling, and based on the historical data set, use multiple linear regression or random forest algorithm to train the weight coefficient w i , in order to minimize the calibration error, there is:

[0099] min∑(BP true -(BP raw -∑w i ·ΔE i )) 2

[0100] ΔE i = E i,current - E i,baseline

[0101] Among them, BP true is the blood pressure calibration value collected regularly, BP raw is the original collected blood pressure value, ΔE i is the error between the actual and ideal environmental parameters, E i,current is the actual environmental parameter, E i,baseline is the ideal environmental parameter;

[0102] Since the screening basis of each variable in the data set is mainly based on its potential impact on blood pressure. For environmental parameters, such as temperature, humidity, air pressure, noise, light intensity, etc., which are related to blood pressure, they are retained. For other possible interfering variables, if their correlation with blood pressure is not significant after statistical analysis, or their occurrence frequency is extremely low and the impact is extremely small in the actual measurement scenario, they are excluded. For example, some environmental factors unique to certain special geographical locations that do not appear in the general usage scenarios can be excluded from the variables;

[0103] At this time, in the specific implementation process, the specific operation process of using multiple linear regression to train the weight coefficient includes: setting a random small value as the initial regression coefficient, and using the gradient descent method for iterative optimization. By calculating the gradient of the loss function with respect to the regression coefficient and updating the coefficient along the opposite direction of the gradient until the loss function converges to the minimum value, so as to determine the optimal weight coefficient w i ; ​

[0104] At this time, in the specific implementation process, the specific operation process of training the weight coefficient using the random forest algorithm includes: determining the number of decision trees according to the scale of the data set and computing resources, then constructing and training multiple decision trees through sampling with replacement of the samples, and finally obtaining the optimal weight coefficient w for the final prediction based on the results of the decision trees. i This method can enhance the stability and generalization ability of the model.

[0105] a3. During the blood pressure calibration and verification process, dynamically adjust the calibration amplitude k according to the individual age differences of the users whose blood pressure is measured to obtain the optimized blood pressure value BP. adj There is:

[0106]

[0107] Among them, age represents age, and β is the attenuation factor, which is obtained by fitting through the historical data of the users whose blood pressure is measured.

[0108] Furthermore, in terms of dynamically adjusting the calibration amplitude according to the individual differences of users, the adjustment strategies of β values corresponding to different age ranges and vascular elasticity indexes are as follows: For the age factor, generally speaking, as age increases, vascular elasticity gradually decreases, and the sensitivity of blood pressure to environmental influence will change. The age can be divided into multiple intervals, such as under 30 years old, 30 - 50 years old, and over 50 years old. For people under 30 years old, their physical functions are relatively good, vascular elasticity is strong, and the influence of the environment on blood pressure is relatively small. At this time, the β value can be set to a relatively small value, such as 0.8; for people aged 30 - 50 years old, the blood vessels begin to show a certain degree of aging, the environmental influence gradually increases, and the β value can be adjusted to 1.2; for people over 50 years old, the vascular elasticity significantly decreases, and they are more sensitive to environmental changes, and the β value can be set to 1.5.

[0109] For the vascular elasticity index, it can be obtained through professional medical detection means, such as measuring the elastic modulus of blood vessels by vascular ultrasound examination. If the elastic modulus is at the upper limit of the normal range, it indicates that the vascular elasticity is good, and the β value can be appropriately reduced; if the elastic modulus is at the lower limit of the normal range, it indicates that the vascular elasticity is poor, and the β value should be increased accordingly. Through this refined adjustment strategy, it can more accurately reflect the individual differences of different users and achieve more accurate blood pressure calibration.

[0110] At this time, by fusing multi-source data and performing dynamic calibration processing in the data processing and control module, it is possible to comprehensively analyze multi-dimensions by combining environmental parameters and user physiological signals. At the same time, through multiple linear regression, random forest algorithm, and age difference dynamic adjustment strategy, the original blood pressure value can be accurately calibrated, thereby improving the reliability and personalized adaptability of blood pressure data, and finally realizing the precision and scientific nature of blood pressure management.

[0111] (c) A data analysis and prediction unit for analyzing and extracting specified information from environmental data and blood pressure data, and performing deep learning in combination with historical data to predict blood pressure changes.

[0112] Furthermore, the analysis process includes: detecting the trend of blood pressure fluctuations, determining whether it is affected by environmental factors, evaluating whether the user's blood pressure is within the healthy range, calculating potential health risks, etc. In addition, based on the learning of historical data, the system can predict the future blood pressure changes of the user. For example, if the temperature rises sharply or the air quality deteriorates, the system will analyze how these changes affect the user's blood pressure level and give a health warning in advance. The prediction model may be continuously optimized through the training data set, and the algorithm parameters are adjusted through real-time data.

[0113] Among them, the specific operation process of the data analysis and prediction unit for deep learning includes:

[0114] b1. Based on the time-series environmental data E t , the time-series corrected blood pressure value BP adj,t and the profile data containing the user's health information (including information such as age, medical history, medication records, etc.), construct an LSTM prediction model;

[0115] b2. The input layer of the LSTM prediction model receives the time-series data window [E t-n , BP adj,t-n , …, E t , BP adj,t and performs sliding window sampling

[0116] When performing sliding window sampling, the window size is selected as 6 (i.e., the data of the past 1 hour) based on the consideration of the characteristics of blood pressure changes and data correlation. Blood pressure changes have a certain continuity and trend in a short period of time. A time span of 1 hour can capture the recent blood pressure fluctuations and will not cause data redundancy and excessive computational burden due to too large a window. At the same time, through experimental comparison, it is found that with a window size of 6, the model achieves a better balance in capturing the blood pressure change trend and prediction accuracy;

[0117] b3. Set two stacked LSTM layers in the LSTM prediction model, with 64 neurons in each layer, and add Dropout = 0.2 to prevent overfitting;

[0118] b4. Output the blood pressure prediction value at the future time point t + 1 through the fully connected layer And optimize the environmental adjustment strategy through the auxiliary fitness function.

[0119] Specifically, the specific training process of the LSTM prediction model includes:

[0120] Data augmentation is performed on historical data. The sliding window sampling method is adopted, and Gaussian noise with σ = 1 mmHg is added to improve the generalization ability of the model. At this time, the loss function Γ is specifically as follows:

[0121]

[0122] Among them, BP pred,i is the predicted blood pressure value at the i-th time, and BP ture,i is the blood pressure calibration value collected regularly at the i-th time. λ is the intensity adjustment value of L2 regularization;

[0123] For the user's health record, the underlying parameters of the LSTM are frozen, and only the top fully connected layer is trained to make the model more suitable for the individual situation of the user;

[0124] In summary, the auxiliary function is obtained:

[0125]

[0126] Among them, BP adj is the optimized blood pressure value, BP true is the blood pressure calibration value collected regularly, ΔE i is the error between the actual and ideal environmental parameters, and BP target is the target blood pressure range set by the user;

[0127] In the auxiliary function, |BP adj -BP ture |, ∑ i |ΔE i |, |BP pred -BP target | represent measurement error, environmental adjustment cost, and trend deviation respectively. λ1 = 0.6, λ2 = 0.3, λ3 = 0.1 (weights are configurable) represent the adjustable influence factors (weights) of measurement error, environmental adjustment cost, and trend deviation respectively, and λ1 + λ2 + λ3 = 1.

[0128] At this time, by constructing a deep learning model based on LSTM in the data analysis and prediction unit and combining with the auxiliary fitness function to optimize the environmental adjustment strategy, the trend of blood pressure change can be predicted and the measurement error, environmental adjustment cost, and trend deviation can be balanced, so as to improve the foresight and resource utilization efficiency of health management, and ultimately help users achieve proactive blood pressure risk prevention and control.

[0129] After data analysis, the decision support and control instruction output function makes specific decisions based on the analysis results. For example, if the system predicts that the user's blood pressure may rise, or certain factors in the environment (such as excessive noise, low humidity, etc.) may have an adverse effect on blood pressure, the control module will issue corresponding control instructions according to the pre-set health standards and thresholds. These instructions may include sending a signal to the blood pressure measurement module to request a re-measurement to confirm the result, or sending a reminder notification to the user interaction module to inform the user of the measures to be taken, such as adjusting environmental parameters or taking a rest. At the same time, the decision support module can also be linked with other devices such as a smart home system, automatically adjusting the settings of devices such as an air purifier, air conditioner, and humidifier, so as to achieve all-round health management.

[0130] Coordinating and communicating with other modules is also a key function of the data processing and control module. The system needs to ensure seamless connection with the environmental perception module, blood pressure measurement and user interaction module, ensuring smooth and efficient data transfer and sharing. During the entire operation of the system, the data processing and control module is responsible for coordinating the flow, processing, and feedback of data, ensuring the synchronous operation of each module. For example, when the blood pressure measurement data is abnormal, the control module will immediately start an analysis program to evaluate whether it is necessary to change the environmental settings or notify the user for further examination, ensuring that the system can respond in a timely manner.

[0131] In addition, to ensure the long-term stability and accuracy of the system, the data processing and control module needs to have the ability of self-learning and optimization. Based on the continuously collected real-time data and historical data, the module can improve the accuracy of data analysis and prediction through algorithm iteration and optimization. Machine learning algorithms can be continuously updated to gradually improve the system's understanding of the relationship between the environment and blood pressure, so as to make more accurate health management decisions.

[0132] Replaceable components of the data processing and control module include different models of microprocessors (such as ARM, DSP, etc.) and memories (such as Flash, RAM, etc.). This module receives data from the environmental perception module, processes it through pre-set algorithms, and then transmits the processing results to the blood pressure measurement and user interaction module to achieve precise control of blood pressure measurement.

[0133] (3) Blood pressure measurement and interaction module, which is used to perform blood pressure measurement operations and feedback the measurement results to the user. In addition, this module is connected to a blood pressure monitor to obtain the user's blood pressure data in real time and upload these data to the data processing and control module for analysis.

[0134] Meanwhile, the user interface in the module will provide feedback on health information to the user according to the analysis results, including the real-time value of blood pressure, trend analysis, and warning prompts. To improve the user experience, the module may also include a voice assistant, a touch screen display, charts, and a notification system, etc., to facilitate the user to more intuitively understand their health status. The replaceable components can be a sphygmomanometer sensor, a touch screen display, a voice recognition module, etc.

[0135] In summary, the above modules cooperate with each other to form a closed-loop system to ensure the accuracy and reliability of blood pressure measurement results in a changing environment.

[0136] On the other hand, the present invention also discloses a blood pressure management method, which uses any one of the above blood pressure management systems based on dynamic environment perception to perform operations, including:

[0137] Real-time monitoring and regulation of environmental information through the environment perception module, and then transmitting the data to the data processing and control module;

[0138] The data processing and control module adjusts the blood pressure measurement strategy according to the environmental data, and then sends instructions to the blood pressure measurement and interaction module;

[0139] The blood pressure measurement and interaction module controls the execution of blood pressure measurement operations and feeds back the results to the user.

[0140] On yet another aspect, the present invention also discloses a blood pressure management device, which is characterized in that it includes an interaction interface, a processor, a sphygmomanometer main body, a cuff, and a measurement sensor, wherein:

[0141] The interaction interface, the sphygmomanometer main body, and the measurement sensor are all connected to the processor;

[0142] The interaction interface and the cuff are both arranged on the sphygmomanometer main body, and the sphygmomanometer main body and the measurement sensor are used to contact the user to perform blood pressure measurement operations.

[0143] The processor includes a memory and a processor. When the computer program stored in the memory is executed by the processor, the processor performs the method steps of blood pressure management and detection based on any one of the above systems.

[0144] At this time, the blood pressure measurement part is mainly composed of a sphygmomanometer main body, a cuff, and a measurement sensor, and the user interaction part includes a display screen, buttons, a voice prompt device, etc. The replaceable components include sphygmomanometer main bodies of different brands, different types of sensors (such as pressure sensors, photoelectric sensors, etc.), and different forms of interaction devices (such as touch screens, voice recognition systems, etc.).

[0145] The blood pressure measurement and user interaction module is closely connected to the environmental perception module and the data processing and control module. It measures blood pressure according to the instructions of the data processing and control module, displays the results to the user, and at the same time receives the user's operation instructions to achieve human-computer interaction.

[0146] Therefore, by integrating an interactive interface, a processor, and sensors in the blood pressure management device and equipping it with an adaptive algorithm and an environmental regulation linkage mechanism, it can provide real-time, accurate, and personalized blood pressure monitoring and environmental optimization functions, thereby enhancing the convenience and effectiveness of user health management and ultimately significantly improving the quality of healthy life of hypertensive patients.

[0147] On the other hand, the present invention also discloses a computer-readable storage medium storing a computer program. When the computer program is executed by a processor, the processor is caused to execute the steps of the above method.

[0148] In another embodiment provided by the present application, there is also provided a computer program product containing instructions. When it runs on a computer, the computer is caused to execute the corresponding blood pressure management method through any of the blood pressure management systems based on dynamic environmental perception in the above embodiments.

[0149] It can be understood that the system provided by the embodiments of the present invention corresponds to the method provided by the embodiments of the present invention. The explanations, examples, and beneficial effects of related content can refer to the corresponding parts in the above method.

[0150] The embodiments of the present application also provide an electronic device, including a processor, a communication interface, a memory, and a communication bus. Among them, the processor, the communication interface, and the memory complete communication with each other through the communication bus.

[0151] The memory is used to store a computer program.

[0152] The processor is used to execute the program stored on the memory and implement the corresponding blood pressure management method through the above blood pressure management system based on dynamic environmental perception.

[0153] The communication bus mentioned in the above electronic device can be a peripheral component interconnect standard bus or an extended industry standard architecture bus, etc. This communication bus can be divided into an address bus, a data bus, a control bus, etc.

[0154] The communication interface is used for communication between the above electronic device and other devices.

[0155] The memory can include a random access memory and can also include a non-volatile memory, such as at least one disk memory. Optionally, the memory can also be at least one storage device located far from the aforementioned processor.

[0156] The above-mentioned processor may be a general-purpose processor, including a central processing unit, a network processor, etc.; it may also be a digital signal processor, an application-specific integrated circuit, a field-programmable gate array, or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components.

[0157] It should also be noted that the electronic device also includes a terminal device, which can also be referred to as a terminal, a user equipment, a mobile station, a mobile terminal, etc. The terminal device can be a mobile phone, a smart TV, a wearable device, a tablet computer, a computer with wireless transceiver function, a virtual reality terminal device, an augmented reality terminal device, a wireless terminal in industrial control, a wireless terminal in unmanned driving, a wireless terminal in remote surgery, a wireless terminal in smart grid, a wireless terminal in transportation safety, a wireless terminal in smart city, a wireless terminal in smart home, and so on. The embodiments of the present application do not limit the specific technologies and specific device forms adopted by the terminal device.

[0158] In the above embodiments, it can be implemented in whole or in part by software, hardware, firmware, or any combination thereof. When implemented using software, it can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the processes or functions described in the embodiments of the present application are generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from a website, a computer, a server, or a data center to another website, a computer, a server, or a data center in a wired manner (such as coaxial cable, optical fiber, digital subscriber line) or a wireless manner (such as infrared, wireless, microwave, etc.). The computer-readable storage medium can be any available medium that can be accessed by a computer, or a data storage device such as a server, a data center, etc. that includes one or more available media integrated. The available medium can be a magnetic medium (such as a floppy disk, a hard disk, a magnetic tape), an optical medium (such as a DVD), or a semiconductor medium (such as a solid-state drive SSD), etc.

[0159] The above are only the preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

[0160] In addition, it should be noted that if there are directional indications (such as up, down, left, right, front, back...) involved in the embodiments of the present invention, the directional indications are only used to explain the relative positional relationship and movement conditions between components in a certain specific posture. If the specific posture changes, the directional indications will also change accordingly.

[0161] In addition, if there are descriptions such as "first" and "second" involved in the embodiments of the present invention, the descriptions of "first", "second", etc. are only for descriptive purposes and cannot be understood as indicating or implying their relative importance or implicitly indicating the quantity of the indicated technical features. Thus, the features defined with "first" and "second" may explicitly or implicitly include at least one such feature. In addition, the meaning of "and / or" appearing throughout the text includes three parallel scenarios. Taking "A and / or B" as an example, it includes Scenario A, or Scenario B, or the scenario where both A and B are satisfied simultaneously. In addition, in the embodiments of the present invention, "a plurality of" means two or more. In addition, the technical solutions between various embodiments can be combined with each other, but it must be based on the ability of those of ordinary skill in the art to implement. When the combination of technical solutions results in contradictions or cannot be implemented, it should be considered that such a combination of technical solutions does not exist and is not within the scope of protection required by the present invention.

Claims

1. A blood pressure management system based on dynamic environment perception, characterized in that Including: An environmental perception module, which is used to monitor environmental parameters including temperature, humidity, air pressure, and noise in real time, and through multi-parameter coupling compensation, link an external environmental regulation system to optimize and adjust the measurement environment to a specified state; A data processing and control module, which is used to deeply process and analyze the collected data, and make control decisions based on the analysis results to achieve precise management and monitoring of blood pressure; A blood pressure measurement and interaction module, which is used to perform blood pressure measurement operations and feedback the measurement results to the data processing and control module and the user.

2. The blood pressure management system based on dynamic environment perception according to claim 1, wherein The regulation logic of the multi-parameter coupling compensation of the dynamic environmental perception module to link and regulate the external environment includes: When the temperature > 28°C and the humidity > 80%, a comprehensive compensation strategy of using the vasodilation compensation factor γ1 × the blood viscosity correction factor γ2 is adopted to compensate the blood pressure measurement result δ0, and the blood pressure measurement result δ is obtained l It is: δ l = δ0 × γ1 × γ2; In the state where the noise > 70 dB and the light > 1500 lux, activate the dual-channel filtering algorithm to eliminate artifact signals; Construct an environmental compensation system, and based on the ISO 81060-3 standard certification, generate an environmental interference index EEI in real time. When EEI > 0.85, automatically trigger an environmental regulation instruction, and link an external environmental regulation system to adjust the measurement environment to the specified state.

3. The blood pressure management system based on dynamic environment perception according to claim 1, wherein Inside the data processing and control module, there are: A data reception and preprocessing unit, which is used to receive real-time data from the environmental perception module and the blood pressure measurement and interaction module, and perform data preprocessing operations; A data processing and control unit for multi-source data fusion and correction of environmental parameters and physiological data, dynamically perceiving environmental parameters and user physiological signals to dynamically correct the original blood pressure value, thereby eliminating environmental interference and outputting an optimized blood pressure value BP adj , and constructing an adaptive algorithm framework based on Kalman filtering to achieve noise suppression through a state space model; A data analysis and prediction unit, which is used to analyze and extract specified information from environmental data and blood pressure data, and perform deep learning in combination with historical data to predict blood pressure changes.

4. The blood pressure management system based on dynamic environment perception according to claim 3, characterized in that, The specific operation process of the data processing and control unit for multi-source data fusion and correction includes: a1. By dynamically sensing environmental parameters including temperature, humidity, air pressure, noise, and light, an environmental feature matrix E within a time period t is generated t , there is: E t = [T t , H t , P t , N t , L t ​ Among them, T t is the temperature parameter, H t is the humidity parameter, P t is the air pressure parameter, N t is the noise parameter, L t is the light parameter; a2. Perform environmental impact factor modeling and, based on the historical data set, use multiple linear regression or random forest algorithm to train the weight coefficient w i , to minimize the correction error, there is: min∑(BP true -(BP raw -∑w i ·ΔE i )) 2 ΔE i = E i,current - E i,baseline Among them, BP true is the blood pressure calibration value collected regularly, BP raw is the original blood pressure value collected, ΔE i is the error between the actual and ideal environmental parameters, E i,current is the actual environmental parameter, E i,baseline is the ideal environmental parameter; a3. Dynamically adjust the correction amplitude k according to the individual age difference of the user whose blood pressure is measured during blood pressure calibration and verification to obtain the optimized blood pressure value BP adj , there is: Among them, age represents age, and β is an attenuation factor, which is obtained by fitting the historical data of the blood pressure measured user.

5. The blood pressure management system based on dynamic environment perception according to claim 4, wherein The specific operation process of using multiple linear regression to train the weight coefficient in step a2 includes: Set a random small value as the initial regression coefficient, and use the gradient descent method for iterative optimization. By calculating the gradient of the loss function with respect to the regression coefficient and updating the coefficient in the opposite direction of the gradient until the loss function converges to the minimum value, the optimal weight coefficient w is determined i 。 6. The blood pressure management system based on dynamic environment perception according to claim 4, wherein The specific operation process of using the random forest algorithm to train the weight coefficient in step a2 includes: Determine the number of decision trees according to the dataset scale and computing resources, then construct and train multiple decision trees through sampling with replacement of the samples, and finally obtain the optimal weight coefficient w of the final prediction from the results of the decision trees i 。 7. The blood pressure management system based on dynamic environment perception according to claim 3, characterized in that The specific operation process of the data analysis and prediction unit for deep learning includes: b1. Based on the time-series environmental data E t and the time-series corrected blood pressure value BP adj,t as well as the profile data containing the user's health information, construct an LSTM prediction model; The input layer of the b2.LSTM prediction model receives the time series data window [E t-n , BP adj,t-n , …, E t , BP adj,t , and performs sliding window sampling, where the window size n is set to 6 to represent the data of the past 1 hour; b3. Set two stacked LSTM layers in the LSTM prediction model, with 64 neurons in each layer, and add Dropout = 0.2 to prevent overfitting; b4. Output the predicted blood pressure value BP at the future time point t+1 through the fully connected layer pred , and optimize the environmental regulation strategy through the auxiliary fitness function.

8. The blood pressure management system based on dynamic environment perception according to claim 7, wherein The specific training process of the LSTM prediction model includes: Adopt the method of sliding window sampling to perform data augmentation on historical data, and add Gaussian noise σ = 1 mmHg to improve the generalization ability of the model. At this time, the loss function Γ is specifically: Regularization Among them, BP pred,i is the predicted blood pressure value at the i-th time, and BP ture,i is the calibrated blood pressure value regularly collected at the i-th time, and λ is the intensity adjustment value of L2 regularization; For the user's health record, freeze the parameters of the bottom layer of the LSTM and only train the top fully connected layer to obtain an auxiliary function: Among them, BP adj is the optimized blood pressure value, BP true is the blood pressure calibration value collected regularly, ΔE i is the error between the actual and ideal environmental parameters, BP target is the target blood pressure range set by the user; |BP in the auxiliary function adj -BP ture |, ∑ i |ΔE i |, |BP pred -BP target | represent measurement error, environmental adjustment cost, and trend deviation respectively, and λ1, λ2, and λ3 represent the adjustable influence factors of measurement error, environmental adjustment cost, and trend deviation respectively, and λ1 + λ2 + λ3 = 1.

9. A blood pressure management method, which uses the blood pressure management system based on dynamic environment perception described in any one of claims 1-8 above to perform operations, characterized in that, Including: Real-time monitor the environmental information through the environmental perception module and regulate the environment, and then transmit the data to the data processing and control module; The data processing and control module adjusts the blood pressure measurement strategy according to the environmental data, and then sends the instruction to the blood pressure measurement and interaction module; The blood pressure measurement and interaction module controls the execution of blood pressure measurement operations and feedbacks the results to the user.

10. A blood pressure management device, characterized in that, Including an interaction interface, a processor, a blood pressure meter main body, a cuff, and a measurement sensor, where: The interaction interface, the blood pressure meter main body, and the measurement sensor are all connected to the processor; The interaction interface and the cuff are both provided on the main body of the sphygmomanometer, and the main body of the sphygmomanometer and the measurement sensor are used to contact the user to perform blood pressure measurement operations. The processor includes a memory and a processor. The memory stores a computer program. When the computer program is executed by the processor, the processor is caused to perform the method steps of blood pressure management and detection based on the system according to any one of claims 1 to 8.

Images