Omnibearing intelligent cardiopulmonary health management system and method and electronic equipment

Abstract

The invention relates to the technical field of health management and medical information, and discloses an all-dimensional intelligent cardiopulmonary health management system and method and electronic equipment, and the system comprises a biological sensing module which is used for continuously collecting key physiological indexes of the heart rate, the oxyhemoglobin saturation, the respiratory rate and the motion state of a user; the data processing module is used for storing the collected physiological data and historical exercise records of the user through a central database of a cloud computing platform; the deep learning analysis module is a prediction model built based on a deep learning framework and is used for analyzing the physiological data of the user and generating a personalized exercise plan; the real-time monitoring module is used for detecting the physiological state of the user and automatically adjusting the exercise intensity or type when the physiological state exceeds a preset threshold value. The technical scheme that the multi-mode biosensor is combined with the real-time monitoring module is adopted, and the technical effect of comprehensively monitoring the cardiopulmonary function state of the user in real time is achieved.

Description

Technical Field

[0001] The present invention relates to the fields of health management and medical information technology, and specifically to an all-round intelligent cardiopulmonary health management system, method and electronic device. Background Art

[0002] With the increasing emphasis on personal health management, especially in the field of cardiopulmonary health, a variety of health monitoring devices and systems have emerged on the market. These technical solutions usually rely on biosensors to collect basic physiological data (such as heart rate, blood oxygen saturation, respiratory rate, etc.) and provide health advice for users by analyzing this data. However, despite the gradual increase in the popularity of these devices and systems, there are still many deficiencies in the existing technologies, which restrict the further development of health management systems in terms of personalization and real-time performance.

[0003] Firstly, existing health management systems generally use single or a few sensors for data collection, such as optical heart rate sensors or blood oxygen monitoring sensors. Although this design can meet the basic needs during static monitoring, it often faces problems of incomplete data collection dimensions and insufficient accuracy in dynamic scenarios (such as during user exercise). For example, a single sensor is difficult to accurately capture the multi-dimensional physiological state of a user during intense exercise. Especially when the optical heart rate sensor is interfered by intense exercise, its measurement results are prone to deviation, resulting in data that cannot truly reflect the actual health status of the user. In addition, the relationship between exercise state data (such as step frequency, acceleration) and physiological indicators cannot be associated through a simple collection mechanism, thus limiting the comprehensiveness and accuracy of the data.

[0004] Secondly, most existing health management systems only provide generalized exercise advice, usually relying on preset fixed templates. These templates are usually based on the user's gender, age or simple health goals, lacking in-depth analysis of individual differences. For example, when a user with relatively weak cardiopulmonary function uses a fixed template for high-intensity exercise, it may cause the body to operate with overload, posing potential safety hazards. For users with better physical fitness, they may not be able to achieve the desired fitness goals due to insufficient training intensity. This simplistic advice generation method is difficult to dynamically adjust the training plan according to the user's real-time physiological state and long-term health goals, resulting in a lack of pertinence and scientific nature in the exercise plan.

[0005] Third, the real-time monitoring ability of existing systems is significantly insufficient. Although some devices can provide continuous monitoring of physiological data such as heart rate and respiratory rate, the data is mostly simply judged based on fixed thresholds, ignoring the personalized needs of users and the rapid changes in physiological states during exercise. For example, during high-intensity exercise, data such as the user's heart rate and blood oxygen saturation may quickly approach the dangerous range, but due to the lack of dynamic prediction and intelligent judgment capabilities, traditional systems are unable to issue warnings in a timely manner or adjust the exercise plan, which is extremely likely to increase the risks during exercise. In addition, the real-time monitoring function often only stays at the level of data recording and cannot achieve intelligent analysis and intervention of the user's real-time state.

[0006] Finally, the data feedback and user interaction functions of the health management system are also relatively weak, making it difficult for users to intuitively understand their own health status or training effects. Most devices in the existing technologies feedback data in simple numerical or text forms, and it is difficult for users to intuitively understand the trends and meanings of data changes from this information. For example, users may only see the fluctuations of heart rate values but cannot judge the changes in cardiopulmonary function or the effects of exercise plans from them. In addition, the lack of interactive and customizable feedback methods also reduces users' interest and stickiness in using the system.

[0007] In summary, the main deficiencies of existing health management systems are mainly reflected in the limitations in the data collection dimension, the lack of personalized analysis capabilities, the lag in real-time monitoring and dynamic adjustment, and the lack of user interaction and feedback mechanisms. These problems make it difficult for the system to meet the actual needs of users in different scenarios and complex health goals. Especially in the field of cardiopulmonary health management, the shortcomings of insufficient system personalization level and lack of real-time adjustment ability are more prominent, and urgent improvements are needed to better adapt to the diverse needs and dynamic states of users.

[0008] Therefore, the present invention proposes an all-round intelligent cardiopulmonary health management system, method and electronic device to solve the deficiencies of the existing technologies. Summary of the Invention

[0009] Aiming at the deficiencies of the existing technologies, the present invention provides an all-round intelligent cardiopulmonary health management system, method and electronic device, which solve the problems of single data collection dimension, insufficient personalized analysis, lag in real-time monitoring and lack of user interaction.

[0010] To achieve the above purposes, the present invention is realized through the following technical solutions: An all-round intelligent cardiopulmonary health management system, including: A biosensing module: used for continuously collecting key physiological indicators of the user's heart rate, blood oxygen saturation, respiratory rate and exercise state; Data processing module: Stores the collected physiological data and the user's historical exercise records through the central database of the cloud computing platform; Deep learning analysis module: A prediction model built based on the deep learning framework, used to analyze the user's physiological data and generate personalized exercise plans; Real-time monitoring module: Used to detect the user's physiological state and automatically adjust the exercise intensity or type when the physiological state exceeds the preset threshold; User interface module: Provides an intuitive and friendly user interface for displaying health reports, exercise progress, and recovery suggestions.

[0011] Preferably, the biosensing module includes multimodal sensors, and the sensors include an optical heart rate sensor, a gas exchange measurement component, and an accelerometer, which are used to capture the user's cardiopulmonary function state in multiple dimensions.

[0012] Preferably, the deep learning analysis module uses a long short-term memory network and a support vector machine model to predict the user's cardiopulmonary endurance limit and fatigue threshold.

[0013] Preferably, the user interface module includes: Display module, used to display the user's personalized health data and training progress; Prompt module, used to push the user's training intensity adjustment and rest suggestions.

[0014] The present invention also provides a comprehensive intelligent cardiopulmonary health management method, including the following steps: Collect the user's heart rate, blood oxygen saturation, respiratory rate, and exercise status through the biosensing module; Upload the collected data to the central database of the cloud computing platform and store it; Analyze the user data using the deep learning framework to generate a personalized exercise prescription; Real-time monitor the user's physiological state, and when it is detected that the user's state exceeds the threshold, adjust the exercise intensity and type; Push exercise suggestions, health reports, and recovery guidance to the user through the user interaction module.

[0015] Preferably, the personalized exercise prescription includes a training plan designed according to the user's maximum oxygen uptake and resting heart rate.

[0016] Preferably, it also includes dynamically optimizing the selection of the training scenario by combining external environment data, and the external environment data includes weather conditions and air quality index. The specific steps are as follows: Obtain the real-time weather data of the user's location, including temperature, humidity, and precipitation; Retrieve the air quality index data and determine whether the external environment is suitable for outdoor exercise; Generate suitable training scenario recommendations based on external environmental data combined with the user's current physiological state; When the external environment is not conducive to outdoor sports, automatically switch to a suitable indoor training plan and adjust the training intensity and type.

[0017] The present invention also provides an all-round intelligent cardiopulmonary health management electronic device, which includes: An intelligent wearable device integrating a biosensor module and a real-time monitoring module; A mobile terminal, communicating with the intelligent wearable device, for displaying the user's health data and exercise suggestions.

[0018] Preferably, the intelligent wearable device includes an embedded microcontroller and a storage module for locally storing the user's key health data and synchronizing it with the cloud.

[0019] Preferably, the mobile terminal includes an AI coach module for generating real-time training guidance and rest suggestions based on the user's historical physiological feedback.

[0020] The present invention provides an all-round intelligent cardiopulmonary health management system, method and electronic device. It has the following beneficial effects: 1. The present invention adopts the technical solution of combining multi-modal biosensors with a real-time monitoring module, achieving the technical effect of comprehensively and real-time monitoring the user's cardiopulmonary function state. Compared with the existing technology that relies on a single sensor to collect a single physiological data, it solves the problems of single data collection dimension and insufficient accuracy, and can dynamically adapt to the complex physiological changes in different exercise scenarios.

[0021] 2. The present invention generates exercise prescriptions through a deep learning analysis module combined with the user's personalized data, realizing high-precision training plan customization. Compared with the existing technology that mainly uses general exercise templates, it avoids the problem of low training effect caused by failure to consider individual differences, and better meets the user's need for precise health management.

[0022] 3. The present invention uses a real-time status monitoring and dynamic threshold adjustment mechanism to effectively ensure the safety of the user during exercise. Most of the existing technologies use fixed thresholds for judgment, which are difficult to cope with the rapid changes in the state during exercise, while the present invention solves this deficiency and can timely adjust the training intensity or type to reduce the exercise risk caused by over-fatigue or environmental factors.

[0023] 4. The present invention provides visual health reports and recommendation push through the user interaction interface module, achieving clear display of health data and personalized interaction. Compared with the prior art solutions with single and unintuitive data feedback forms, the present invention significantly improves the user's ability to understand their health status, enabling users to intuitively grasp their exercise effects and health trends, and at the same time providing a more user-friendly operation experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] Figure 1 is the system architecture diagram of the present invention; Figure 2 is the method flow chart of the present invention; Figure 3 is the device diagram of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0025] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0026] Please refer to Figure 1 , the embodiments of the present invention provide a full-range intelligent cardiopulmonary health management system, including: Biological sensing module: used to continuously collect key physiological indicators of the user's heart rate, blood oxygen saturation, respiratory rate, and exercise status; Data processing module: stores the collected physiological data and the user's historical exercise records through the central database of the cloud computing platform; Deep learning analysis module: a prediction model built based on the deep learning framework, used to analyze the user's physiological data and generate personalized exercise plans; Real-time monitoring module: used to detect the user's physiological state, and automatically adjust the exercise intensity or type when the physiological state exceeds the preset threshold; User interaction interface module: provides an intuitive and friendly user interface for displaying health reports, exercise progress, and recovery suggestions.

[0027] Bio - sensing Module To achieve comprehensive management of the user's cardiopulmonary health, the bio - sensing module of the present invention undertakes the crucial task of collecting physiological data. This module collaborates with other modules to form a complete dynamic health management system. Specifically, the bio - sensing module not only collects conventional indicators such as the user's heart rate, blood oxygen saturation, and respiratory rate, but also captures the exercise state and other bio - signals related to cardiopulmonary function through components such as accelerometers and gas exchange measurements. These data are then uploaded by the data - processing module to the central database, providing basic support for subsequent analysis. When designing this module, the present invention fully considers the diversity of sensors, the real - time nature of data collection, and its dynamic adaptability to the user's state.

[0028] In this embodiment, the bio - sensing module includes multi - modal sensors. These sensors work together based on different physical principles and can collect the user's multi - dimensional physiological data with high precision.

[0029] Specifically, the module contains an optical heart - rate sensor. Generally, the optical heart - rate sensor is based on photoplethysmography (PPG). By irradiating the skin with a light source and measuring the change in the intensity of the reflected light, the heart - rate signal is calculated. As an option, in some embodiments, the light source can be a green LED light source with a light wavelength between 500nm and 600nm to optimize the signal detection accuracy under different skin - color conditions. In addition, the collection of the reflected light is completed by a high - sensitivity photodetector, and its output signal is pre - processed by a band - pass filter to eliminate the interference of ambient light noise.

[0030] The gas - exchange measurement component is an important part of this module and is used to evaluate the user's respiratory rate and metabolic state. Specifically, in some embodiments, the gas - exchange measurement component utilizes the changes in the oxygen concentration and carbon - dioxide concentration in the exhaled gas. Through the formula: where, is the oxygen consumption, is the carbon - dioxide production, is the minute ventilation volume, F IO2 and F ICO2 are the oxygen and carbon - dioxide fractions in the inhaled gas respectively, and F EO2 and F ECO2 are the oxygen and carbon - dioxide fractions in the exhaled gas respectively. By calculating the user's oxygen consumption and carbon - dioxide emission in real - time, this component can deduce the user's respiratory rate and metabolic rate, especially suitable for respiratory monitoring during high - intensity exercise.

[0031] In some embodiments, an accelerometer is integrated into the module to monitor the user's motion state. Generally, the accelerometer captures motion trajectory data through a triaxial acceleration sensor, such as step frequency, speed, and attitude angle changes. Specifically, the module calculates the motion acceleration through the following formula: where a x 、a y 、a z are the acceleration components in three directions respectively. By combining the acceleration with data such as heart rate and breathing frequency, the user's exercise intensity and exercise type can be accurately judged.

[0032] To achieve synchronous acquisition of multimodal data, the biosensing module designs a data fusion processing unit. In one possible implementation, this unit aligns the data from different sensors based on timestamps and uses a weighted fusion algorithm to optimize the accuracy of multi-source data. The formula of the weighted fusion algorithm is as follows: where D fusion is the fused data result, D i is the acquisition data of the i-th sensor, and w i is the weight coefficient, and the weight coefficient is dynamically adjusted according to the reliability of the sensor. For example, when it is detected that the signal of the optical heart rate sensor is disturbed due to the user's intense exercise, the system will reduce its weight and increase the weight of the accelerometer data.

[0033] In some embodiments, to further improve the comfort of the user's wearing and the stability of the sensing data, the hardware shell of the biosensing module is made of flexible material, and a shock-absorbing structure is designed inside to reduce the influence on the sensor signal during motion. At the same time, a conductive coating is covered on the surface of the module to shield electromagnetic interference.

[0034] As an option, the biosensing module can also support the wireless charging function, making it more suitable for the needs of long-term wearing. In one possible implementation, the power management chip of the module can dynamically adjust the sampling frequency of the sensor. For example, it reduces the sampling frequency in the resting state to save power, and automatically switches to the high-frequency sampling mode when detecting high-intensity exercise.

[0035] Generally speaking, the biosensing module of the present invention can achieve high-precision, high-real-time, and multi-scenario physiological data acquisition by integrating multimodal sensors and combining optimized data fusion and preprocessing algorithms. This module provides a reliable data basis for subsequent deep learning analysis and health management, and at the same time has strong scalability and adaptability, laying a solid technical foundation for realizing intelligent health management.

[0036] The data processing module is designed to achieve efficient storage and subsequent processing of the data collected by the biosensing module in the present invention. The data processing module works closely with the biosensing module to form a complete health data management link. Specifically, the data processing module receives multi-modal physiological data uploaded by the biosensing module, formats, cleans, stores, and manages it, providing stable and high-quality data support for the subsequent deep learning analysis module. The core objective of this module is to ensure the real-time, integrity, and accuracy of data processing, guaranteeing the scientific nature of subsequent analysis.

[0037] In this embodiment, the functions of the data processing module mainly include data uploading, data cleaning, data storage, and management. To implement these functions, multiple subsystems and algorithms are designed inside the module.

[0038] Generally, the biosensing module transmits the collected physiological data to the data processing module in real time through wireless communication methods (such as Bluetooth, Wi-Fi, or ZigBee). As an option, in some embodiments, the data processing module can achieve short-distance data synchronization through near-field communication (NFC) or low-power Bluetooth (BLE), which is suitable for the application requirements of low-power scenarios.

[0039] After receiving the data, the data processing module will first perform formatting processing on the data. Specifically, the data formats generated by different sensors may vary. For example, heart rate data may be a continuous time series, while acceleration data may contain multiple three-dimensional coordinate vectors. To achieve unified processing, the data processing module converts these raw data into a unified time series data structure: D(t) = {x 1 (t), x 2 (t),..., x n (t)} where D(t) represents the data set at time point t, and x 1 (t), x 2 (t), etc. are the output values of different sensors.

[0040] After completing the data formatting, the module will perform data cleaning. Data cleaning is an important function of the data processing module, aiming to eliminate outliers and noise to ensure the accuracy and reliability of the data. In some embodiments, the cleaning process includes the following steps: Outlier detection: By setting reasonable threshold ranges, detect and remove physiological data that exceeds the normal range. For example, if the heart rate signal suddenly jumps to an unphysiological level (such as greater than 250 bpm or less than 30 bpm), it will be marked as an outlier and removed.

[0041] Noise filtering: In a possible implementation, the data processing module applies a band-pass filter to the received signal to remove high-frequency and low-frequency noise. For heart rate signals, the commonly used filter frequency range is 0.5 Hz to 3 Hz.

[0042] Data interpolation: To handle the problem of missing time-series data due to signal loss in the sensor, this embodiment adopts a linear interpolation algorithm: where x(t 1 ) and x(t 2 ) are the valid data at time points t 1 and t 2 respectively, and x(t) is the interpolation result at time point t.

[0043] The cleaned data will be classified and stored in the central database. The central database adopts a distributed architecture, which can handle data upload requests from multiple users simultaneously and support cross-device data synchronization. In a possible implementation, the database uses a NoSQL database (such as MongoDB) to store unstructured data and supports efficient retrieval through an SQL query interface. Specifically, the module creates a unique identifier (UUID) for each user and organizes the historical data records of the user with the timestamp as the key index.

[0044] To ensure data storage security and privacy protection, the data processing module encrypts all uploaded data. In some embodiments, the AES-256 encryption algorithm is used to perform end-to-end encryption on the transmitted data, and combined with an authentication mechanism to ensure that only authorized users can access the data.

[0045] In terms of management, the data processing module designs a multi-layer caching strategy to improve the data access speed. Generally, the module will save the recently used high-frequency data in the local cache and store the historically accessed low-frequency data in the cloud. In a possible implementation, the local cache adopts a circular queue structure to ensure the maximization of memory utilization.

[0046] To improve data transmission efficiency, the data processing module also designs a data compression function. Specifically, the module uses a lossless compression algorithm (such as LZW or Huffman coding) to compress and store redundant data and decompresses it during data transmission. In this way, the occupancy of network bandwidth is significantly reduced.

[0047] As an option, the data processing module can also dynamically adjust the data storage strategy. For example, when the device battery is low, the module can delay data upload and preferentially save it to local storage; when the device is charging or connected to a stable network, it will upload it to the cloud in batches.

[0048] In some embodiments, to enhance the intelligence of the module, the data processing module can combine the user's usage habits to intelligently predict possible types of accessed data. For example, for users with frequent movement, the module will preferentially load movement-related data; for the resting monitoring scenario, the module will pay more attention to the continuity of heart rate and blood oxygen data.

[0049] In summary, the data processing module in this embodiment can comprehensively process the data collected by the biosensing module through a reasonable design process and efficient algorithms, ensuring the accuracy, reliability, and real-time nature of the data. This not only provides high-quality input for subsequent in-depth learning analysis but also provides users with long-term, traceable health records, meeting various needs of intelligent health management.

[0050] For the in-depth learning analysis module to achieve an in-depth understanding of the user's health status and the precise generation of exercise plans, the in-depth learning analysis module utilizes the high-quality physiological data provided by the data processing module to construct an artificial intelligence-based prediction model. Through the comprehensive analysis of multi-dimensional data such as heart rate, blood oxygen saturation, respiratory rate, and exercise status, this module can generate personalized exercise prescriptions and real-time predict the user's fatigue status and cardiopulmonary endurance limit. This module is the core part of the entire system, and its design fully considers data complexity and user individual differences to ensure the scientific nature and effectiveness of the recommended results.

[0051] In this embodiment, the main functions of the in-depth learning analysis module include feature extraction, model training, personalized exercise prescription generation, and real-time status prediction. To achieve these functions, the module adopts a variety of deep learning algorithms and frameworks.

[0052] Generally, feature extraction is the first step of the in-depth learning analysis module. The goal of feature extraction is to extract representative numerical features from the original data to provide input for subsequent model training. In a possible implementation, the module performs windowing processing on time series data and extracts statistical features within each time period, such as mean, variance, maximum value, minimum value, and rate of change. For example, for heart rate data, the rate of change can be calculated using the following formula: where ΔHR represents the heart rate change rate, HR(t) and HR(t - Δt) represent the heart rate values at time points t and t - Δt respectively, and Δt is the time interval. Specifically, when extracting respiratory rate features, a spectral analysis method can be used. The time-domain signal is converted to the frequency domain through the fast Fourier transform (FFT), and the dominant frequency component is identified to calculate the respiratory rate: where f breathis the respiratory rate, and FFT(signal) is the frequency domain representation of the signal.

[0053] After feature extraction, the deep learning analysis module trains a model on the extracted features. The core of the model training is to establish an individualized prediction model for the user to evaluate the cardiorespiratory endurance limit and fatigue state. In some embodiments, the module adopts a combined model of long short-term memory network (LSTM) and support vector machine (SVM). Specifically, LSTM is suitable for processing time series data. Through the design of memory units and forget gates, its structure can capture long-term dependencies. The calculation formula of LSTM is as follows: f t = σ(W f · [h t-1 , x t + b f ) i t = σ(W i · [h t-1 , x t + b i ) C t = f t · C t-1 + i t · tanh(W c · [h t-1 , x t + b c ) o t = σ(W o · [h t-1 , x t + b o ) h t = o t · tanh(C t ) where f t is the activation value of the forget gate, used to determine whether to "forget" the memory of the previous time step at the current time step, i t is the activation value of the input gate, used to determine the influence of the current input data on the memory state, o t is the activation value of the output gate, used to determine how the current memory state affects the output result, C t is the memory cell state at the current time step, which is the core part for the network to store long-term dependency information, C t-1 is the memory cell state of the previous time step, representing the memory information from the previous time step, h t is the hidden state at the current time step, representing the network output at the current time step, h t-1is the hidden state of the previous time step and is part of the input at the current time step, x t is the input data at the current time step, representing the external input features passed to the network, W f , W i , W c , W o are the weight matrices for the forget gate, input gate, memory update calculation, and output gate, used to calculate the activation values of different gates, b f , b i , b c , b o are the bias terms for the forget gate, input gate, memory update calculation, and output gate, which are fixed parameters for each gate and used to adjust the calculation results.

[0054] Generally, training an LSTM model requires a large-scale labeled dataset. In one possible implementation, this embodiment uses the collected historical data, combined with the user's subjective feeling score (RPE value) as the labeled data, to train a user-specific fatigue state prediction model.

[0055] When generating a personalized exercise prescription, the module combines the user's health goals (such as fat loss, muscle gain, improving cardiorespiratory endurance) and current state to design a reasonable training plan. As an option, the design of the training plan will determine the training intensity according to the user's maximum oxygen uptake (VO2 max) and resting heart rate through the following formula: HR target = HR rest +(HR max - HR rest )·intensity where, HR target is the target heart rate, HR rest is the resting heart rate, HR max is the maximum heart rate (generally 220 - age), and intensity is the training intensity coefficient (usually between 0.5 and 0.85).

[0056] Real-time state prediction is another important function of the deep learning analysis module. In some embodiments, the module continuously monitors the user's current state, evaluates in real time whether it is close to the fatigue threshold, and adjusts the exercise plan. For example, when it detects that the user's heart rate change rate is too high and the breathing frequency exceeds the normal range, the module will prompt the user to reduce the exercise intensity or stop exercising through the mobile terminal or smart wearable device.

[0057] As an option, the deep learning analysis module can also optimize the prediction results through external environmental data. For example, in a high-temperature or high-humidity environment, the module will automatically reduce the predicted value of the fatigue threshold to avoid discomfort caused by the environmental impact on the user.

[0058] For the real-time monitoring module to achieve dynamic adjustment and safety management of the user's health status, the real-time monitoring module of the present invention closely combines the functions of the aforementioned biosensing module, data processing module, and deep learning analysis module to complete the real-time assessment of the user's physiological state and the intelligent adjustment of the training plan. Specifically, through the real-time data provided by the biosensing module and combined with the prediction results of the deep learning analysis module, this module dynamically monitors whether the user is in a safe exercise state and adjusts the exercise plan or issues a reminder in case of abnormalities.

[0059] In this embodiment, the design of the real-time monitoring module aims to meet the real-time nature of data processing, the timeliness of response, and the stability of the system.

[0060] Generally, the real-time monitoring module receives continuous physiological data inputs from the biosensing module, including heart rate, blood oxygen saturation, and respiratory rate, etc. As an option, in some embodiments, the module can also obtain the user's exercise state data (such as step frequency, acceleration) and external environment data (such as temperature, humidity) as auxiliary monitoring bases.

[0061] Specifically, this module uses a threshold judgment mechanism to quickly evaluate the user's current physiological state. The threshold is set based on the user's personalized physical indicators, such as maximum heart rate (HR max ) and resting heart rate (HR rest ), and combines the following formula to dynamically calculate the target heart rate range: HR low =HR rest +0.5·(HR max -HR rest ) HR high =HR rest +0.85·(HR max -HR rest ) Wherein, HR low and HR high respectively represent the lower and upper limits of the target heart rate interval. The module determines whether the user is within the target training intensity range based on the real-time collected heart rate data. If the heart rate exceeds HR high , the system will trigger a warning and suggest that the user reduce the exercise intensity.

[0062] In a possible implementation manner, this module not only makes judgments based on thresholds but also combines the prediction results of the deep learning analysis module for more accurate state monitoring. For example, when it is detected that the relationship between the heart rate change rate (ΔHR) of the user and the respiratory rate (f breath ) deviates from the normal mode, the system will evaluate whether the user is approaching the fatigue threshold. The specific formula is as follows: Fatigue S core = w 1 ΔHR + w 2 ·f breath Among them, Fatigue_Score is the fatigue score, w 1 and w 2 are weight parameters, ΔHR and f breath are the heart rate change rate and respiratory rate respectively. The module determines whether the user needs to pause training or adjust the intensity according to the set scoring threshold.

[0063] To enhance the flexibility and responsiveness of monitoring, a multi-level feedback mechanism is designed in this embodiment. When it is detected that the user's physiological state is abnormal, the module will notify the user through vibration prompts, sound and light alarms or the mobile terminal. As an option, the system can also directly adjust the user's training plan according to the degree of abnormality. For example, when the user's heart rate exceeds the safe range but does not reach the critical level, the module will recommend that the user reduce the exercise intensity; when the user's state is close to the limit, the module will pause the current training and provide restorative guidance.

[0064] In some embodiments, the real-time monitoring module also optimizes the monitoring results using external environmental data. Specifically, the module will obtain the temperature, humidity and air quality index (AQI) of the environment where the user is located through networked devices and adjust the training safety threshold. For example, in a high-temperature environment, the system will lower the safety upper limit of the heart rate to avoid the risk of heat stroke caused by high temperature for the user. At this time, the adjustment formula for the heart rate threshold is: HR adjusted = HR original ·(1 - k·T factor ) Among them, HR adjusted is the adjusted heart rate threshold, HR original is the original heart rate threshold, k is the environmental impact coefficient, and T factor is the temperature impact factor.

[0065] To improve the adaptability of the monitoring module, a self-learning mechanism is also designed in this embodiment, enabling the system to optimize the monitoring parameters based on the user's long-term data accumulation. Generally, the module will regularly analyze the user's historical training records and dynamically adjust the heart rate threshold and the weight coefficient of the fatigue score to ensure the accuracy and personalization of monitoring.

[0066] As an option, the real-time monitoring module supports linkage with the user interface module to provide a visual real-time status monitoring function. Specifically, users can view their current heart rate, respiratory rate, and fatigue score on a mobile terminal or smart wearable device and adjust their training strategies according to system prompts. This function is particularly applicable to high-intensity training scenarios that require immediate feedback.

[0067] User Interface Module To achieve an intuitive display of users' health data and personalized interactive feedback, the present invention designs a user interface module. This module works closely with the aforementioned real-time monitoring module and deep learning analysis module, responsible for transmitting the processed physiological data, training plans, and health suggestions to users and supporting users to customize and adjust some functions of the system. Through this module, users can grasp their own health status in real time and optimize their behaviors according to system prompts, thereby enhancing the convenience and effectiveness of health management.

[0068] In this embodiment, the core functions of the user interface module include data visualization, training advice push, and user setting adjustment. The design of the module focuses on the intuitiveness, ease of use, and immediacy of the interaction.

[0069] Generally, the user interface module uses the screen of a mobile terminal (such as a smartphone) or a smart wearable device as the display carrier. Specifically, the module displays the users' real-time health data and historical records in a graphical way. For example, the users' heart rate, respiratory rate, and blood oxygen saturation can be displayed in the form of a line graph or a bar graph, which can not only highlight the change trend of the data but also facilitate users' understanding. For example, the time series curve of heart rate data can be standardized according to the following formula to facilitate users to compare the training intensity in different time periods: where HR normalized represents the standardized heart rate value, HR is the current heart rate, and HR rest and HR max are the resting heart rate and the maximum heart rate respectively.

[0070] As an option, this module can adjust the display content according to the target needs of different users. For example, for users who focus on fat loss, the interface will highlight the total calorie consumption value of the training; for users who pay attention to cardiopulmonary endurance, the interface will give priority to showing the trend change of the maximum oxygen uptake (VO2 max).

[0071] In a possible implementation, the user interface module can also push personalized training suggestions generated by the system. Specifically, the training suggestions are based on the output results of the deep learning analysis module and combine the safety tips of the real-time monitoring module to provide targeted guidance information. For example, when the user's fatigue score is close to the threshold, the system will recommend that the user reduce the training intensity or increase the rest time. In addition, the module will automatically generate a training summary after the exercise, including the following content: Average heart rate, maximum heart rate, minimum heart rate; Total training duration, total calories burned; Training objective completion rate and health score.

[0072] This summary can help users comprehensively understand the effect of this training and provide a reference for the next stage of the training plan.

[0073] To enhance the user's autonomy, the user interface module in this embodiment supports multiple personalized setting options. Generally, users can set health goals (such as fat loss, muscle gain, or endurance improvement) through the module, and adjust the training intensity or time period. For example, users can set the goal to "burn 3000 kcal per week", and the system will dynamically adjust the content and intensity of each training according to this goal. As an option, users can also set a reminder mechanism. For example, when the exercise intensity is too high or the physiological state is abnormal, the system will push vibration or sound prompts through the interface.

[0074] In some embodiments, the module supports the multi-device linkage function, enabling users to seamlessly view data on different terminals. For example, users can view the real-time heart rate and training tips on a smart bracelet, while viewing detailed health reports and historical data analysis results on a mobile application.

[0075] In the design of data interaction, the module adopts a responsive interface architecture to adapt to display devices of different sizes. Specifically, the interface designs multiple dynamic interaction windows that can automatically adjust the layout and content according to the user's operations. For example, when the user clicks on a certain piece of data (such as a heart rate curve graph), the interface will automatically pop up detailed numerical statistics information and display a trend analysis graph in a slidable manner.

[0076] To ensure the security of user data, the user interface module in this embodiment designs a strict permission management mechanism. Generally, only the user himself or authorized devices can access health data. In a possible implementation, the module combines biometric technologies (such as fingerprint recognition or face recognition) to verify the user's identity to enhance the strength of data privacy protection.

[0077] In addition, this module also supports docking with external data platforms, such as third-party health management applications or medical service institutions. Through the API interface, users can synchronize their health data to other platforms for more comprehensive health management.

[0078] Please refer to Figure 2 , the present invention also provides a full-range intelligent cardiopulmonary health management method, including the following steps: Collect the user's heart rate, blood oxygen saturation, respiratory rate, and exercise status through the biosensing module; Upload the collected data to the central database of the cloud computing platform and store it; Use the deep learning framework to analyze the user data and generate personalized exercise prescriptions; Monitor the user's physiological status in real time. When it is detected that the user's status exceeds the threshold, adjust the exercise intensity and type; Push exercise suggestions, health reports, and recovery guidance to the user through the user interaction module.

[0079] S1. Collect the user's heart rate, blood oxygen saturation, respiratory rate, and exercise status through the biosensing module. When the user is performing daily activities or exercises, the system continuously collects multiple key physiological indicators through the biosensing module, such as heart rate, blood oxygen saturation, and respiratory rate. At the same time, use the built-in accelerometer and other motion sensors to monitor the user's exercise status, including exercise type, step frequency, acceleration changes, etc. As an option, the system can also combine environmental factors, such as temperature and humidity, to enhance the adaptability of the data. During the collection process, high-precision sensors and multi-modal signal processing technologies are used to ensure the real-time and accuracy of the data.

[0080] S2. Upload the collected data to the central database of the cloud computing platform and store it. The system uploads the collected physiological data to the cloud through wireless transmission methods (such as Bluetooth or Wi-Fi). During the upload process, the data will be automatically classified into different categories (such as timestamp, heart rate, exercise status, etc.) and stored hierarchically in the central database. Generally, the system will use encryption technology to transmit and store the data to ensure the privacy and security of users. As a possible implementation method, the database designs an efficient indexing mechanism to enable users to quickly retrieve and query historical data.

[0081] S3. After the personalized exercise prescription data is generated by analyzing the user data using a deep learning framework, the system comprehensively analyzes the user's historical data and real-time data through the deep learning framework. Specifically, the system uses time series analysis models such as LSTM to extract the changing trends of the user's physiological data, and generates a personalized exercise prescription in combination with the user's health goals (such as fat loss and muscle gain). As an option, the exercise prescription will include detailed suggestions on exercise type, intensity, duration, and rest intervals. For example, for users with a relatively low maximal oxygen uptake, the system may recommend mainly medium- to low-intensity aerobic training and gradually increase the training intensity.

[0082] S4. During the user's exercise, the system continuously monitors the user's physiological state through the real-time monitoring module and makes judgments based on the preset personalized safety thresholds. If it is detected that indicators such as the user's heart rate, blood oxygen saturation, or respiratory rate exceed the set range, the system will immediately re-evaluate the data through the deep learning analysis module and adjust the exercise plan. For example, it is recommended that the user reduce the current exercise intensity, extend the rest time, or switch to low-intensity recovery training. Generally, the response time during the monitoring process will be optimized to ensure the immediacy and safety of the system.

[0083] S5. After completing the data analysis and monitoring, the system pushes the results to the user in an intuitive way through the user interaction module. These contents include exercise suggestions (such as adjusting the current training plan), health reports (such as the recent changing trend of the heart rate), and recovery guidance (such as stretching movements and rest suggestions). Generally, this information will be presented in a graphical way. For example, a line chart shows the heart rate change, and a pie chart shows the completion of the training goals. As an option, the system also supports voice prompts or vibration reminders to facilitate the user to receive key information during exercise, and at the same time allows the user to set personalized preferences in the interface, such as training goals and notification types.

[0084] Please refer to Figure 3 , the present invention also provides an all-round intelligent cardiopulmonary health management electronic device, and the electronic device includes: An intelligent wearable device integrating a biosensor module and a real-time monitoring module; A mobile terminal, communicating with the intelligent wearable device, for displaying the user's health data and exercise suggestions.

[0085] An electronic device is used to collect users' physiological data in real time, monitor users' exercise status, and provide personalized health management services for users through data analysis and an interaction interface. The electronic device includes a smart wearable device and a mobile terminal, which are linked through wireless communication technology to provide users with comprehensive health data monitoring and exercise guidance.

[0086] The smart wearable device integrates a biosensor module and a real-time monitoring module to collect key physiological data such as users' heart rate, blood oxygen saturation, breathing frequency, and exercise status in real time and perform preliminary processing. Generally, the smart wearable device adopts a lightweight and comfortable design, suitable for users to wear in various scenarios, such as daily life, fitness training, or professional sports. In a possible implementation, the smart wearable device also includes sensor modules such as high-precision optical sensors, gas exchange measurement components, and triaxial accelerometers to ensure the comprehensiveness and accuracy of physiological data collection.

[0087] As an option, the smart wearable device also has an embedded processor and a local storage module for filtering, compressing, and storing the collected raw data. If the communication is interrupted or the data cannot be uploaded to the cloud in real time, the device will store the temporary data locally and automatically synchronize it to the mobile terminal or cloud platform after the network is restored. In addition, the smart wearable device supports wireless charging function for convenient long-term use, and its shell is waterproof, capable of meeting the usage requirements of users in various environments.

[0088] The mobile terminal is connected to the smart wearable device through wireless communication methods (such as Bluetooth or Wi-Fi) and is used to receive and process the data uploaded from the wearable device. The main functions of the mobile terminal include displaying users' health data, generating exercise reports, pushing training suggestions, and providing personalized recovery guidance. For example, the mobile terminal will present users' physiological data in a visual form, including heart rate change curves, blood oxygen level trend charts, and breathing frequency distribution maps. In addition, the mobile terminal will push a training plan and recovery suggestions suitable for the current state to the user according to the exercise prescription generated by the deep learning analysis module.

[0089] Generally, the mobile terminal provides users with convenient health management functions through an intuitive and user-friendly interaction interface. In a possible implementation, users can adjust training goals, set safety thresholds, or customize reminder methods through the mobile terminal. For example, users can set the upper and lower limits of the heart rate. When the actual heart rate exceeds the range, the terminal will issue an alarm reminder and synchronously display the adjusted exercise suggestions.

[0090] In addition, the mobile terminal also supports two-way communication with the cloud database, enabling users to access their historical health data and long-term trend reports anytime and anywhere. For example, users can view their training results in the past week or month in the application and obtain the health scores and improvement suggestions generated by the system. As an option, the mobile terminal can also be docked with third-party health management platforms or medical institutions through the API interface to facilitate data sharing or receiving professional guidance for users.

[0091] Through the collaborative work of the smart wearable device and the mobile terminal, the electronic device of the present invention can build a complete cardiopulmonary health management system, comprehensively improving the user's health management experience from data collection, analysis to personalized feedback.

[0092] Although the embodiments of the present invention have been shown and described, it will be understood by those of ordinary skill in the art that various changes, modifications, substitutions and variations can be made to these embodiments without departing from the principles and spirit of the present invention, and the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. Comprehensive intelligent cardiopulmonary health management system, characterized by: include: Biosensor module: used to continuously collect key physiological indicators of the user's heart rate, blood oxygen saturation, respiratory rate and exercise status; Data processing module: stores collected physiological data and user historical exercise records through the central database of the cloud computing platform; Deep learning analysis module: A prediction model built based on a deep learning framework to analyze the user's physiological data and generate a personalized exercise plan; Real-time monitoring module: used to detect the user's physiological state and automatically adjust the intensity or type of exercise when the physiological state exceeds the preset threshold; User Interaction Interface Module: Provides an intuitive and friendly user interface for displaying health reports, exercise progress and recovery suggestions.

2. The all-round intelligent cardiopulmonary health management system according to claim 1, characterized in that: The biosensing module includes a multimodal sensor, which includes an optical heart rate sensor, a gas exchange measurement component and an accelerometer, and is used to capture the user's cardiopulmonary function status in multiple dimensions.

3. The all-round intelligent cardiopulmonary health management system according to claim 1, characterized in that: The deep learning analysis module uses a long short-term memory network and a support vector machine model to predict the user's cardiopulmonary endurance limit and fatigue threshold.

4. The all-round intelligent cardiopulmonary health management system according to claim 1, characterized in that: The user interaction interface module comprises: Display module, used to display the user's personalized health data and training progress; The prompt module is used to push training intensity adjustments and rest suggestions to users.

5. A comprehensive intelligent cardiopulmonary health management method, applied to the comprehensive intelligent cardiopulmonary health management system according to any one of claims 1 to 4, characterized in that: The following steps are involved: The user's heart rate, blood oxygen saturation, breathing rate and movement status are collected through the biosensor module; Upload the collected data to the central database of the cloud computing platform and store it; Use deep learning framework to analyze user data and generate personalized exercise prescriptions; Monitor the user's physiological state in real time, and adjust the intensity and type of exercise when it is detected that the user's state exceeds the threshold; Exercise suggestions, health reports and recovery guidance are pushed to users through the user interaction module.

6. The all-round intelligent cardiopulmonary health management method according to claim 5, characterized in that: The personalized exercise prescription includes a training plan designed according to the user's maximum oxygen uptake and resting heart rate.

7. The all-round intelligent cardiopulmonary health management method according to claim 5, characterized in that: It also includes the selection of the training scenario dynamically optimized in combination with external environmental data, wherein the external environmental data includes weather conditions and air quality index, and the specific steps are as follows: Get real-time weather data for the user's location, including temperature, humidity, and precipitation; Retrieve air quality index data to determine whether the external environment is suitable for outdoor sports; Generate appropriate training scenario recommendations based on external environment data and the user's current physiological state; When the external environment is not conducive to outdoor exercise, it automatically switches to a suitable indoor training plan and adjusts the training intensity and type.

8. An all-round intelligent cardiopulmonary health management electronic device, applied to the all-round intelligent cardiopulmonary health management method as claimed in any one of claims 5 to 7, characterized in that: The electronic device comprises: Smart wearable devices, integrating biosensor modules and real-time monitoring modules; The mobile terminal communicates with the smart wearable device and is used to display the user's health data and exercise suggestions.

9. The all-round intelligent cardiopulmonary health management system, method and electronic device according to claim 8, characterized in that: The smart wearable device includes an embedded microcontroller and a storage module, which are used to locally store the user's key health data and synchronize it with the cloud.

10. The all-round intelligent cardiopulmonary health management electronic device according to claim 8, characterized in that: The mobile terminal includes an AI coaching module for generating real-time training guidance and rest suggestions based on the user's historical physiological feedback.

Images