Health data analysis method and device based on Internet of Things, equipment and medium

Abstract

The invention relates to the field of artificial intelligence, can be applied to business system platforms of financial science and technology, medical health and the like, and discloses a health data analysis method, device, equipment and medium based on the Internet of Things, and the method comprises the steps: collecting multi-dimensional health sensing data and environment data through a plurality of sensors; performing data fusion processing on the multi-dimensional health sensing data to obtain multi-modal fusion sensing data; performing adaptive data calibration on the multi-modal fusion sensing data according to the environment data to obtain calibrated sensing data; and inputting the calibration sensing data into a pre-trained dynamic early warning model in the local Internet of Things equipment, and outputting a health prompt after performing health dynamic early warning analysis on the calibration sensing data locally. Data calibration is carried out on the multi-modal fusion sensing data through the environment data to improve accuracy and reliability, dynamic early warning analysis is carried out locally based on the calibration data, the risk of privacy disclosure is avoided, and safe and efficient health analysis and early warning processing are achieved.

Description

Technical Field

[0001] The present invention relates to the field of computer technology, and in particular to a health data analysis method, device, equipment and medium based on the Internet of Things. Background Art

[0002] IoT-based health data analysis is the process of collecting, transmitting, storing, and analyzing individual or group health data in real time through sensors, smart devices, and cloud computing technologies to support health management decisions. Due to the widespread application of health data in various scenarios, health data analysis is playing an increasingly important role in different fields.

[0003] For example, in the healthcare sector, elderly people can use wearable devices (such as smart watches) and medical terminals (such as portable monitors) to collect real-time physiological indicators such as heart rate, blood pressure, and blood sugar, analyze the data, and then issue health warnings or generate personalized health recommendations (such as chronic disease management plans, rehabilitation training guidance, etc.). For another example, in the financial sector, after users wear wearable devices to collect and analyze health data, the analysis results can provide reference information on health status for financial insurance companies to conduct risk assessments or adjust insurance premiums.

[0004] Currently, health data is usually collected based on consumer-grade devices such as health bracelets, which has the problem of insufficient accuracy. In addition, the data usually needs to be uploaded to the cloud for analysis and processing, which may not only cause delays but also pose the risk of privacy leakage. Summary of the Invention

[0005] In view of the above-mentioned deficiencies in the prior art, the purpose of the present invention is to provide a health data analysis method, device, equipment and medium based on the Internet of Things that can be applied to the medical field, financial technology or other related fields. Its main purpose is to improve the accuracy and reliability of health data analysis and realize safe and efficient health analysis and early warning processing.

[0006] The technical solutions of the present invention are as follows:

[0007] A first aspect of the present invention provides a health data analysis method based on the Internet of Things, comprising:

[0008] Collect multi-dimensional health sensor data and environmental data through multiple sensors in local IoT devices;

[0009] Performing data fusion processing on the multi-dimensional health sensor data to obtain multimodal fusion sensor data;

[0010] Performing adaptive data calibration on the multimodal fusion sensor data according to the environmental data to obtain calibrated sensor data;

[0011] The calibration sensor data is input into a dynamic early warning model pre-trained in the local Internet of Things device, and a corresponding health prompt is output after a health dynamic early warning analysis is performed on the calibration sensor data locally.

[0012] A second aspect of the present invention provides a health data analysis device based on the Internet of Things, comprising:

[0013] A data acquisition module is used to collect multi-dimensional health sensor data and environmental data through multiple sensors in local IoT devices;

[0014] A data fusion module is used to perform data fusion processing on the multi-dimensional health sensor data to obtain multimodal fusion sensor data;

[0015] a data calibration module, configured to perform adaptive data calibration on the multimodal fusion sensor data according to the environmental data to obtain calibrated sensor data;

[0016] The health dynamic analysis module is used to input the calibration sensor data into the dynamic warning model pre-trained in the local Internet of Things device, and output corresponding health prompts after performing health dynamic warning analysis on the calibration sensor data locally.

[0017] A third aspect of the present invention provides a computer device comprising at least one processor; and

[0018] a memory communicatively connected to the at least one processor; wherein,

[0019] The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the above-mentioned health data analysis method based on the Internet of Things.

[0020] A fourth aspect of the present invention provides a non-volatile computer-readable storage medium, which stores computer-executable instructions. When the computer-executable instructions are executed by one or more processors, the one or more processors can execute the above-mentioned health data analysis method based on the Internet of Things.

[0021] Beneficial effects: The present invention discloses a health data analysis method, apparatus, device and medium based on the Internet of Things. Compared with the existing technology, the embodiment of the present invention collects multi-dimensional health sensor data and environmental data through multiple sensors in a local Internet of Things device; performs data fusion processing on the multi-dimensional health sensor data to obtain multimodal fusion sensor data; performs adaptive data calibration on the multimodal fusion sensor data according to the environmental data to obtain calibrated sensor data; inputs the calibrated sensor data into a dynamic early warning model pre-trained in the local Internet of Things device, and outputs corresponding health prompts after performing health dynamic early warning analysis on the calibrated sensor data locally. Data calibration of the multimodal fusion sensor data is performed through environmental data to improve the accuracy and reliability of health data analysis, and dynamic early warning analysis is performed locally based on the calibration data to avoid privacy leakage risks and achieve safe and efficient health analysis and early warning processing. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] In order to more clearly illustrate the solutions in the present invention, a brief introduction is given below to the drawings required for use in describing the embodiments of the present invention. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0023] Figure 1 A schematic diagram of an application environment of the Internet of Things-based health data analysis method provided by an embodiment of the present invention;

[0024] Figure 2 A flow chart of a health data analysis method based on the Internet of Things provided by an embodiment of the present invention;

[0025] Figure 3 A flowchart of step S202 in the health data analysis method based on the Internet of Things provided by an embodiment of the present invention;

[0026] Figure 4 A flowchart of step S203 of the Internet of Things-based health data analysis method provided in an embodiment of the present invention;

[0027] Figure 5 A flowchart of step S204 in the health data analysis method based on the Internet of Things provided by an embodiment of the present invention;

[0028] Figure 6 A schematic diagram of the functional modules of a health data analysis device based on the Internet of Things provided by an embodiment of the present invention;

[0029] Figure 7 A schematic diagram of the hardware structure of a computer device provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0030] To make the objectives, technical solutions, and effects of the present invention more clear and distinct, the present invention is further described in detail below. It should be understood that the specific embodiments described herein are merely for the purpose of explaining the present invention and are not intended to limit the present invention. The embodiments of the present invention are described below with reference to the accompanying drawings.

[0031] The health data analysis method based on the Internet of Things provided by the embodiment of the present invention can be applied in the following areas: Figure 1 In an application environment, the system includes a first terminal device 101, a second terminal device 102, a third terminal device 103, a network 104, and a server 105. The network 104 is a medium for providing a communication link between the first terminal device 101, the second terminal device 102, the third terminal device 103, and the server 105. The network 104 may include various connection types, such as wired and / or wireless communication links, etc.

[0032] The user may use the first terminal device 101, the second terminal device 102, and the third terminal device 103 to interact with the server 105 via the network 104 to receive or send messages, etc. Various communication client applications may be installed on the first terminal device 101, the second terminal device 102, and the third terminal device 103, such as knowledge reading applications, web browser applications, search applications, instant messaging tools, email clients, and / or social platform software (for example only).

[0033] The first terminal device 101 , the second terminal device 102 , and the third terminal device 103 may be various electronic devices having display screens and supporting web browsing, including but not limited to smart phones, tablet computers, laptop computers, desktop computers, and the like.

[0034] The server 105 may be a server that provides various services, such as a backend server that provides support for the content browsed by the user using the first terminal device 101, the second terminal device 102, and the third terminal device 103 (for example only). The backend server may analyze and process the received user requests and other data, and feed back the processing results (such as web pages, information, or data obtained or generated according to the user request) to the terminal device. The server 105 may be a cloud server, also known as a cloud computing server or cloud host, which is a host product in the cloud computing service system to solve the defects of difficult management and weak business scalability in traditional physical hosts and VPS services ("Virtual Private Server", or "VPS" for short). The server 105 may also be a server for a distributed system, or a server combined with a blockchain.

[0035] It should be noted that the health data analysis method based on the Internet of Things provided in the embodiment of the present application can generally be executed by the first terminal device 101, the second terminal device 102, or the third terminal device 103. Accordingly, the health data analysis device based on the Internet of Things provided in the embodiment of the present invention can also be set in the first terminal device 101, the second terminal device 102, or the third terminal device 103. Alternatively, the health data analysis method based on the Internet of Things provided in the embodiment of the present invention can generally be executed by the server 105. Accordingly, the health data analysis device based on the Internet of Things provided in the embodiment of the present invention can generally be set in the server 105.

[0036] It should be understood that the numbers of the above terminal devices, networks and servers are merely illustrative and any number of terminal devices, networks and servers may be provided as required.

[0037] like Figure 2 As shown, the health data analysis method based on the Internet of Things provided by the embodiment of the present invention specifically includes the following steps:

[0038] S201. Collect multi-dimensional health sensor data and environmental data through multiple sensors in local IoT devices.

[0039] In this embodiment, the local IoT device refers to a device that is connected via a network and can collect and transmit data. Specifically, it can be a device pre-installed in a designated location such as a user's home, or it can be a wearable device, etc., and a variety of sensors are embedded in the wearable device, including a PPG sensor (for heart rate and blood oxygen monitoring), an ECG sensor (for heart rhythm monitoring), an accelerometer (for motion status monitoring), an ambient temperature sensor, and a humidity sensor, etc. After the device is started, it uses multiple sensors to collect multi-dimensional data to obtain multi-dimensional health sensor data including physiological data (such as heart rate, blood oxygen) and motion data (such as gait), etc., and also obtains the current user's environmental data such as temperature, humidity, etc. The comprehensiveness of the data is ensured by multi-sensor collection, and it can monitor physiological and environmental factors at the same time, and analyze health data in a timely and accurate manner to capture sudden health events.

[0040] For example, in daily life, users can collect heart rate and heart rhythm data in real time through the PPG and ECG sensors on the smart bracelet they wear, collect acceleration data through the accelerometer, and monitor the indoor temperature through environmental sensors, providing a comprehensive data analysis basis for subsequent health analysis.

[0041] S202: Perform data fusion processing on the multi-dimensional health sensor data to obtain multimodal fusion sensor data.

[0042] In this embodiment, based on the collected multi-dimensional health sensor data, such as PPG data, ECG data, and acceleration data, data from different models are fused. PPG data can monitor heart rate fluctuations, ECG data can verify arrhythmias, and acceleration data can identify motion states to avoid interference from motion artifacts. Multimodal fused sensor data is obtained through data fusion, and functional complementarity between different modalities is achieved to address the problem of insufficient sensor accuracy in existing consumer-grade devices. Specifically, a multi-sensor data fusion algorithm, such as Kalman filtering or machine learning fusion algorithm, can be used to fuse PPG, ECG, and accelerometer data, integrating data from different types of sensors, reducing the error of a single sensor, and improving data accuracy and reliability.

[0043] For example, elderly users wear smart bracelets to monitor their heart health status. In the smart bracelet, the heart rate fluctuation data monitored by the PPG sensor is integrated with the arrhythmia data verified by the ECG sensor, and the motion status information recognized by the accelerometer is combined to avoid the interference of motion artifacts on heart rate monitoring, thereby facilitating a more accurate judgment of the user's heart health status.

[0044] S203 : Performing adaptive data calibration on the multimodal fusion sensing data according to the environmental data to obtain calibrated sensing data.

[0045] In this embodiment, the collected environmental data, such as temperature and humidity, are analyzed to determine their impact on health monitoring data. For example, vasoconstriction in a low-temperature environment may cause unstable heart rate monitoring signals, etc. The multimodal fusion sensor data collected and fused by the sensor is dynamically adjusted according to the environmental parameters to compensate for the impact of environmental changes on the measurement results. For example, when the ambient temperature is low, the heart rate monitoring data can be compensated according to a pre-calibrated calibration strategy, or the sensitivity of the sensor or the data processing threshold can be adjusted to reduce the impact of temperature on heart rate monitoring. Adaptive calibration reduces the impact of environmental factors on the data, making the calibrated data closer to the actual health status and enhancing the accuracy and reliability of the data.

[0046] For example, when a user is at home and the ambient temperature sensor detects that the indoor temperature has dropped below 10°C, adaptive calibration is performed to calibrate the multimodal fusion sensor data based on the calibration rules collected and calibrated within the standard temperature range to ensure that the heart rate and heart rhythm monitoring data remain accurate and reliable in a low temperature environment.

[0047] S204: Input the calibration sensor data into a dynamic early warning model pre-trained in the local IoT device, perform health dynamic early warning analysis on the calibration sensor data locally, and output corresponding health prompts.

[0048] In this embodiment, a pre-trained dynamic early warning model, such as one based on machine learning or deep learning, is loaded onto a local IoT device. Calibrated sensor data is then fed into the model to perform dynamic health early warning analysis. This means that after acquiring the calibration sensor data, all health data analysis is performed locally on the IoT device, avoiding the risk of privacy breaches associated with uploading data to the cloud for health analysis.

[0049] The calibrated sensor data is input into the model, and the model performs dynamic health warnings on the calibrated sensor data based on the pre-learned health warning thresholds for the user in different states. For example, the model will dynamically adjust the warning threshold according to the user's resting heart rate and daily activity pattern, so that dynamic health data analysis can be performed through the health warning threshold that matches the user's current state to confirm the user's health status and output corresponding health prompts. Specifically, when the model detects an abnormal health condition, such as an abnormally high heart rate or arrhythmia, a health prompt message is output on the local IoT device, such as a vibration prompt or voice prompt: "Your heart rate is abnormally high, please rest immediately", etc., to achieve timely and accurate health warning prompts. In this embodiment, all analysis is completed on the local device, reducing dependence on the cloud, improving real-time and privacy protection, and localized analysis also ensures the real-time nature of the warning, and can detect anomalies and issue health prompts in a short time.

[0050] In the above embodiment, the present invention discloses a health data analysis method based on the Internet of Things, which collects multi-dimensional health sensor data and environmental data through multiple sensors in a local Internet of Things device; performs data fusion processing on the multi-dimensional health sensor data to obtain multimodal fusion sensor data; performs adaptive data calibration on the multimodal fusion sensor data according to the environmental data to obtain calibration sensor data; inputs the calibration sensor data into a dynamic early warning model pre-trained in the local Internet of Things device, and outputs corresponding health prompts after performing health dynamic early warning analysis on the calibration sensor data locally. Data calibration of the multimodal fusion sensor data is performed through environmental data to improve the accuracy and reliability of health data analysis, and dynamic early warning analysis is performed locally based on the calibration data to avoid privacy leakage risks and achieve safe and efficient health analysis and early warning processing.

[0051] In one embodiment, Figure 3 As shown, step S202 includes:

[0052] S301, performing noise reduction filtering and normalization processing on the multi-dimensional health sensor data to obtain multi-dimensional pre-processed sensor data;

[0053] S302, obtaining correlations between pre-processed sensor data of different dimensions according to predefined data fusion rules, and determining fusion weights of different dimensions according to the correlations;

[0054] S303 , performing weighted summation on the multi-dimensional pre-processed sensor data according to fusion weights of different dimensions to obtain multimodal fused sensor data.

[0055] In this embodiment, when fusing multi-dimensional health sensor data, pre-processing processes such as noise reduction filtering and normalization processing are first performed. Noise reduction filtering can use the Kalman filter algorithm to perform noise reduction processing on sensor data such as PPG and ECG to remove noise caused by insufficient sensor accuracy or environmental interference; normalization processing converts data collected by different sensors into the same dimension and range, for example, normalizing heart rate and blood oxygen data to the range of 0 to 1 for subsequent fusion processing.

[0056] Then, the correlation between pre-processed sensor data of different dimensions is obtained according to the pre-defined data fusion rules, where data fusion rules refer to rules defined based on the relationship between data to guide data fusion. For example, if ECG data and PPG data are strongly correlated and must be fused, and health analysis has the highest priority, then the correlation between the two is high and a higher weight should be assigned; or if ECG data and positioning sensor data are weakly correlated and can be selectively fused, and health analysis has a low priority, then the correlation between the two is low and a lower weight should be assigned, and so on. In this way, the fusion weight can be dynamically adjusted according to the actual relationship between the data, improving the flexibility of data fusion. According to the determined fusion weight, the pre-processed sensor data of different dimensions are weighted and summed, and the result of the weighted summation is used as multimodal fusion sensor data for subsequent health analysis, which can effectively integrate data of different dimensions and improve the accuracy of the fusion results.

[0057] In one embodiment, Figure 4 As shown, step S203 includes:

[0058] S401, performing environmental feature analysis on the environmental data to extract environmental feature parameters related to health monitoring;

[0059] S402: confirming the user's motion state and body position state according to the multimodal fusion sensing data;

[0060] S403, adaptively matching the environmental characteristic parameters, the user's motion state and body position state according to a preset calibration rule, and determining corresponding adaptive adjustment parameters;

[0061] S404: Perform data calibration on the multimodal fusion sensing data according to the adaptive adjustment parameters to obtain the calibrated sensing data.

[0062] In this embodiment, when calibrating the sensor data based on the environmental data, the environmental characteristics analysis is first performed on the acquired environmental data, that is, the environmental data is analyzed to extract environmental characteristic parameters related to health monitoring, such as the average temperature, the rate of change of humidity, the fluctuation of air pressure, etc. These environmental characteristic parameters have a certain impact on the health monitoring data. For example, low temperature may affect the accuracy of heart rate monitoring, etc. The extraction of environmental characteristic parameters provides a basis for subsequent calibration.

[0063] The system also determines the user's motion and body position based on the multimodal fusion sensor data. For example, it identifies the user's motion state (e.g., stationary, walking, running) based on accelerometer data, and determines the user's body position (e.g., standing, lying down, sitting) by combining accelerometer and gyroscope data. Based on preset calibration rules, the extracted environmental feature parameters are matched with the user's motion and body position. The matching results determine the corresponding adaptive adjustment parameters. These calibration rules guide how to adjust the data based on the environment and state, including the specific adjustment amounts.

[0064] Adaptive adjustment parameters are applied to multimodal fusion sensor data to calibrate the data. For example, when the user is in a low-temperature environment and is stationary, the heart rate data is corrected according to the obtained adaptive adjustment parameters, so that the impact of the environment and status on the health monitoring data can be reasonably compensated. The calibrated data is used as the final calibrated sensor data for subsequent health analysis, reducing the impact of the environment and status on the data and improving the accuracy of the data.

[0065] In one embodiment, Figure 5 As shown, step S204 includes:

[0066] S501: Input the calibration sensor data into a pre-trained dynamic warning model, where the dynamic warning model is obtained by performing federated learning training on the local IoT device;

[0067] S502: Perform health status analysis on the calibration sensor data locally using the dynamic early warning model to determine whether the current health indicator reaches the early warning threshold of the current user;

[0068] S503: If the warning threshold of the current user has been reached, a corresponding warning message is generated and a vibration and / or voice prompt is output through the local Internet of Things device.

[0069] In this embodiment, the calibrated sensor data (such as heart rate, blood oxygen, gait stability, etc.) is input into the dynamic warning model in the local IoT device. The dynamic warning model is obtained by federated learning training on the local IoT device, that is, a personalized dynamic warning model is trained on the local device based on each user's own data, thereby avoiding uploading data to the cloud to protect user data privacy.

[0070] The dynamic warning model performs real-time analysis on the input calibration sensor data, extracts key health indicators (such as heart rate and blood oxygen saturation), and dynamically determines the warning threshold of the current body position or movement state based on the relationship between the user's physical state and health indicators learned during the training phase, and determines whether the current health indicator reaches the dynamically determined warning threshold to achieve dynamic health status analysis. For example, if the user's daily resting heart rate is 60 beats / minute, the model will set the "current heart rate > 85 beats / minute and lasts for 2 minutes" in a calm state after training to trigger a warning, rather than a fixed threshold (such as > 100 beats / minute). This allows for dynamic health analysis based on the user's individual characteristics and status, better adapting to individual differences in users and improving the accuracy of health analysis.

[0071] When a health indicator reaches a warning threshold, a warning message is generated. Specifically, it may include a description of the abnormal indicator and recommended measures. At the same time, based on user preferences or device default settings, a vibration prompt, voice prompt, or a combination of the two is output through the local IoT device to remind the user to take appropriate measures. This embodiment can perform dynamic warning analysis on calibrated health data on the local IoT device and output warning prompts to the user in a timely manner. This not only addresses the privacy leakage risk caused by uploading data to the cloud, but also realizes personalized warnings adapted to the characteristics of different users through the model trained by local federated learning, thereby improving the accuracy and adaptability of dynamic health data warnings.

[0072] In one embodiment, the dynamic early warning model is obtained by performing federated learning training on the local IoT device through the following steps:

[0073] Collecting basic health data and historical sensor data of the user, and constructing local training samples after preprocessing the basic health data and historical sensor data;

[0074] Loading the initialized dynamic warning model, training the dynamic warning model on the local IoT device using the local training samples, updating model parameters and dynamically adjusting the user's warning threshold until a preset convergence condition is met;

[0075] Uploading the local updated parameters of the dynamic warning model to the cloud server, so that the cloud server updates the global model according to the received multiple local updated parameters, and regularly distributes the latest global model parameters to all devices;

[0076] The dynamic early warning model is optimized and updated according to the global model parameters received regularly.

[0077] In this embodiment, a personalized dynamic early warning model is trained through federated learning on the local device to avoid uploading data to the cloud. Specifically, the user's basic health data and historical sensor data are collected from local IoT devices (such as smart bracelets, smart watches, etc.). For example, the level of health data includes age, gender, underlying diseases, etc., and the historical sensor data includes heart rate, blood oxygen, exercise data, etc. The collected data is cleaned, denoised, normalized, and other operations are performed to construct local training samples to ensure data quality. These local training samples can reflect the individual characteristics of different users, thereby supporting the training of personalized models.

[0078] The initialized dynamic warning model is loaded on the local IoT device. Specifically, the parameters of the dynamic warning model can be randomly initialized, or the parameters of the global model can be obtained as initialization parameters by interacting with the cloud server. The initial dynamic warning model is trained using local training samples, and the model parameters are updated through optimization algorithms (such as gradient descent). The warning threshold is dynamically adjusted to adapt to individual differences of users based on the user's historical data and training results. When the model training reaches the preset convergence conditions, such as when the loss function reaches the minimum value or the number of training rounds reaches the upper limit, the training is stopped. Through local data training, the model can adapt to the individual characteristics of the user, improve the accuracy of the warning, and the training data does not need to be uploaded to the cloud, reducing the risk of privacy leakage.

[0079] After the dynamic early warning model training is complete, the local updated parameters obtained from training on the local device are uploaded to the cloud server. The cloud server will receive the local updated parameters from multiple devices and update the global model through an aggregation algorithm, such as using a weighted average strategy to aggregate the updated local model parameters to optimize the global model. At the same time, the updated global model parameters are regularly distributed to all local IoT devices participating in the training. On the local IoT devices, the dynamic early warning model is optimized and updated based on the regularly received global model parameters, allowing the local model to learn more diverse data features and adapt to new data patterns, thereby improving the accuracy of health data analysis and early warning.

[0080] In one embodiment, after step S204, the method further includes:

[0081] When the health reminder includes warning information, identifying the warning event and the corresponding emergency level in the warning information;

[0082] Confirm whether the emergency level of the warning event is higher than the preset level. If higher, output emergency response information to the family end, doctor end and / or hospital end bound to the local Internet of Things device according to the preset cascade response strategy.

[0083] In this embodiment, health alerts may or may not contain warning information based on the user's physical condition. For example, a warning message may be output when the user's health condition is abnormal, while routine health advice may be provided when the user's health condition is normal. When a health alert contains warning information, an automated response mechanism analyzes the health alert and initiates a coordinated response at the corresponding level to promptly intervene in the emergency health abnormality event. Specifically, the warning event and the corresponding emergency level in the warning information can be identified by extracting key information from the warning information, such as abnormal health indicators, to confirm the warning event. The emergency level corresponding to the warning event is then identified according to preset rules. For example, if an abnormal heart rate persists for more than 10 minutes, it is identified as a high-level warning event. The severity of the warning event is divided into levels to respond to warning events of different severities. The identified emergency level of the warning event is compared with the preset level to determine whether a cascade response needs to be triggered. If the emergency level is higher than the preset level, an emergency response message is sent to the family, doctor, and / or hospital according to the preset cascade response strategy.

[0084] Specifically, the preset cascade response strategy is a predefined multi-level response mechanism that triggers different response measures according to the emergency level of the warning event. For example, the emergency level of the event is divided into 10 levels from low to high. If it is higher than level 4, an emergency response message is triggered. Otherwise, only a reminder is given through the local device in the form of vibration or sound and light prompts. When the emergency level of the warning event is higher than level 4, the emergency response information is output to different bound devices according to the preset cascade response strategy. For example, when it reaches level 5-6, the location information is pushed to the family side and the preset emergency number is dialed. When it reaches level 7-8, the emergency task is automatically dispatched to the nearest doctor side. When it reaches level 9-10, the emergency response event is output to the hospital side at the same time to prepare medical resources in advance. Through the cascade response strategy, multi-level linkage between family members, doctors and hospitals is realized, ensuring that relevant personnel can be notified in a timely manner according to the urgency of the event in an emergency, thereby improving the timeliness of intervention.

[0085] In one embodiment, after step S204, the method further includes:

[0086] When the health reminder contains warning information, the corresponding encrypted metadata is generated according to the warning information and uploaded to the cloud server;

[0087] When receiving a data sharing instruction from another device, the cloud server obtains corresponding zero-knowledge proof verification information from the cloud server and sends it to the requesting device, wherein the zero-knowledge proof verification information is obtained by the cloud server after performing zero-knowledge proof processing based on the encrypted metadata.

[0088] In this embodiment, after completing health data analysis on the local device, encrypted metadata is generated based on the warning information when the health condition is abnormal and uploaded to the cloud server. For example, only the "abnormal event type" is uploaded instead of the original health sensor data. When receiving a data sharing instruction from another device, such as a doctor's device, the cloud server uses zero-knowledge proof (ZKP) technology based on the encrypted metadata to generate corresponding zero-knowledge proof verification information and sends this zero-knowledge proof verification information to the requesting device. Because zero-knowledge proof allows the authenticity of data to be verified without revealing the data itself, it can greatly improve data security during the data sharing process. For example, if a local IoT device detects that a user has sent an abnormal heart rate event and outputs a warning message, it will encrypt the corresponding metadata and upload it to the cloud server. When the doctor's terminal device requests data sharing, it can use zero-knowledge proof processing to obtain verification information ("The user has experienced one abnormal heart rate in the past 24 hours"). This information is sent to the doctor's terminal device without sending the specific value, thereby proving the authenticity of the abnormal heart rate warning event without revealing the specific data content. This protects privacy while allowing other devices to verify the authenticity of the data.

[0089] It should be noted that there is not necessarily a certain order between the above steps. A person skilled in the art can understand, based on the description of the embodiments of the present invention, that in different embodiments, the above steps may have different execution orders, that is, they may be executed in parallel, or may be executed interchangeably, etc.

[0090] Further references Figure 6 , as a response to the above Figure 2 The present invention provides an embodiment of a health data analysis device based on the Internet of Things. Figure 2 Corresponding to the method embodiment shown, the device can be specifically applied to various electronic devices.

[0091] like Figure 6 As shown, the health data analysis device 60 based on the Internet of Things in this embodiment includes:

[0092] The data collection module 601 is used to collect multi-dimensional health sensor data and environmental data through multiple sensors in the local IoT device;

[0093] The data fusion module 602 is used to perform data fusion processing on the multi-dimensional health sensor data to obtain multimodal fused sensor data;

[0094] A data calibration module 603 is configured to perform adaptive data calibration on the multimodal fusion sensor data according to the environmental data to obtain calibrated sensor data;

[0095] The health dynamic analysis module 604 is used to input the calibration sensor data into the dynamic warning model pre-trained in the local Internet of Things device, and output corresponding health prompts after performing health dynamic warning analysis on the calibration sensor data locally.

[0096] The module referred to in the present invention refers to a series of computer program instruction segments that can perform specific functions. It is more suitable for describing the execution process of health data analysis based on the Internet of Things than a program. For the specific implementation of each module, please refer to the corresponding method embodiment above, which will not be repeated here.

[0097] In one embodiment, the data fusion module 602 includes:

[0098] a preprocessing unit, configured to perform noise reduction filtering and normalization processing on the multi-dimensional health sensor data to obtain multi-dimensional preprocessed sensor data;

[0099] A weight confirmation unit, configured to obtain correlations between pre-processed sensor data of different dimensions according to predefined data fusion rules, and determine fusion weights of different dimensions according to the correlations;

[0100] The fusion unit is used to perform weighted summation on the multi-dimensional pre-processed sensor data according to fusion weights of different dimensions to obtain multimodal fused sensor data.

[0101] In one embodiment, the data calibration module 603 includes:

[0102] An environmental analysis unit, configured to perform environmental feature analysis on the environmental data and extract environmental feature parameters relevant to health monitoring;

[0103] a state confirmation unit, configured to confirm the user's motion state and body position state based on the multimodal fusion sensing data;

[0104] An adaptive matching unit, configured to adaptively match the environmental characteristic parameters, the user's motion state and body position state according to a preset calibration rule, and determine corresponding adaptive adjustment parameters;

[0105] A data calibration unit is used to perform data calibration on the multimodal fusion sensing data according to the adaptive adjustment parameter to obtain the calibrated sensing data.

[0106] In one embodiment, the health dynamics analysis module 604 includes:

[0107] An input unit, configured to input the calibration sensor data into a pre-trained dynamic early warning model, wherein the dynamic early warning model is obtained by performing federated learning training on the local IoT device;

[0108] A dynamic analysis unit, configured to locally perform a health status analysis on the calibration sensor data using the dynamic early warning model to determine whether the current health indicator reaches an early warning threshold for the current user;

[0109] The early warning prompt unit is used to generate corresponding early warning information and output vibration and / or voice prompts through the local Internet of Things device if the early warning threshold of the current user has been reached.

[0110] In one embodiment, the apparatus 60 further includes:

[0111] A training sample collection module is used to collect the user's basic health data and historical sensor data, and construct local training samples after pre-processing the basic health data and historical sensor data;

[0112] A training module is used to load the initialized dynamic warning model, train the dynamic warning model on the local IoT device using the local training samples, update model parameters, and dynamically adjust the user's warning threshold until a preset convergence condition is met;

[0113] An interactive synchronization module, configured to upload the local updated parameters of the dynamic early warning model to a cloud server, so that the cloud server performs a global model update based on the received multiple local updated parameters and regularly distributes the latest global model parameters to all devices;

[0114] The optimization and updating module is used to optimize and update the dynamic early warning model according to the global model parameters received regularly.

[0115] In one embodiment, the apparatus 60 further includes:

[0116] An event recognition module, configured to, when the health reminder includes warning information, identify the warning event and the corresponding emergency level in the warning information;

[0117] The event response module is used to confirm whether the emergency level of the warning event is higher than the preset level. If it is higher, the emergency response information is output to the family end, doctor end and / or hospital end bound to the local Internet of Things device according to the preset cascade response strategy.

[0118] In one embodiment, the apparatus 60 further includes:

[0119] An encryption upload module is used to generate corresponding encrypted metadata based on the warning information when the health reminder contains warning information and then upload it to the cloud server;

[0120] The data sharing module is used to obtain corresponding zero-knowledge proof verification information from the cloud server and send it to the requesting device when receiving a data sharing instruction from other devices, wherein the zero-knowledge proof verification information is obtained by the cloud server after performing zero-knowledge proof processing based on the encrypted metadata.

[0121] In the above embodiment, the present invention discloses a health data analysis device based on the Internet of Things, which collects multi-dimensional health sensor data and environmental data through multiple sensors in a local Internet of Things device; performs data fusion processing on the multi-dimensional health sensor data to obtain multimodal fusion sensor data; performs adaptive data calibration on the multimodal fusion sensor data according to the environmental data to obtain calibration sensor data; inputs the calibration sensor data into a dynamic early warning model pre-trained in the local Internet of Things device, and outputs corresponding health prompts after performing health dynamic early warning analysis on the calibration sensor data locally. Data calibration is performed on the multimodal fusion sensor data through environmental data to improve the accuracy and reliability of health data analysis, and dynamic early warning analysis is performed locally based on the calibration data to avoid privacy leakage risks and achieve safe and efficient health analysis and early warning processing.

[0122] Another embodiment of the present invention provides a computer device, such as Figure 7 As shown, the computer device 70 includes:

[0123] One or more processors 701 and memory 702, Figure 7 In the description, a processor 701 is used as an example. The processor 701 and the memory 702 can be connected via a bus or other means. Figure 7 The bus connection is taken as an example.

[0124] The processor 701 is used to complete various control logics of the computer device 70. It can be a general-purpose processor, a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field programmable gate array (FPGA), a single-chip microcomputer, an ARM (Acorn RISC Machine) or other programmable logic device, discrete gate or transistor logic, discrete hardware components or any combination of these components. In addition, the processor 701 can also be any traditional processor, microprocessor or state machine. The processor 701 can also be implemented as a combination of computing devices, for example, a combination of a DSP and a microprocessor, multiple microprocessors, one or more microprocessors combined with a DSP and / or any other such configuration.

[0125] Memory 702, as a non-volatile computer-readable storage medium, can be used to store non-volatile software programs, non-volatile computer-executable programs, and modules, such as program instructions corresponding to the IoT-based health data analysis method in the embodiments of the present invention. Processor 701 executes the non-volatile software programs, instructions, and modules stored in memory 702 to execute various functional applications and data processing of computer device 70, thereby implementing the IoT-based health data analysis method in the above-mentioned method embodiments.

[0126] The memory 702 may include a program storage area and a data storage area, wherein the program storage area may store an operating system and application programs required for at least one function; the data storage area may store data created according to the use of the computer device 70, etc. In addition, the memory 702 may include a high-speed random access memory, and may also include a non-volatile memory, such as at least one disk storage device, a flash memory device, or other non-volatile solid-state storage device. In some embodiments, the memory 702 may optionally include a memory remotely located relative to the processor 701, and these remote memories may be connected to the computer device 70 via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof. One or more units are stored in the memory 702, and when executed by one or more processors 701, the steps of the health data analysis method based on the Internet of Things in any of the above-mentioned method embodiments are performed.

[0127] In the above embodiment, the present invention discloses a computer device that collects multi-dimensional health sensor data and environmental data through multiple sensors in a local Internet of Things device; performs data fusion processing on the multi-dimensional health sensor data to obtain multimodal fusion sensor data; performs adaptive data calibration on the multimodal fusion sensor data based on the environmental data to obtain calibration sensor data; inputs the calibration sensor data into a dynamic early warning model pre-trained in the local Internet of Things device, and performs health dynamic early warning analysis on the calibration sensor data locally and outputs corresponding health prompts. Data calibration is performed on the multimodal fusion sensor data through environmental data to improve the accuracy and reliability of health data analysis, and dynamic early warning analysis is performed locally based on the calibration data to avoid privacy leakage risks and achieve safe and efficient health analysis and early warning processing.

[0128] An embodiment of the present invention provides a non-volatile computer-readable storage medium, which stores computer-executable instructions. When the computer-executable instructions are executed by one or more processors, the steps of the health data analysis method based on the Internet of Things in any of the above method embodiments are executed.

[0129] In the above embodiment, the present invention discloses a non-volatile computer-readable storage medium, which collects multi-dimensional health sensor data and environmental data through multiple sensors in a local Internet of Things device; performs data fusion processing on the multi-dimensional health sensor data to obtain multimodal fusion sensor data; performs adaptive data calibration on the multimodal fusion sensor data based on the environmental data to obtain calibrated sensor data; inputs the calibrated sensor data into a dynamic early warning model pre-trained in the local Internet of Things device, and performs health dynamic early warning analysis on the calibrated sensor data locally to output corresponding health prompts. Data calibration is performed on the multimodal fusion sensor data through environmental data to improve the accuracy and reliability of health data analysis, and dynamic early warning analysis is performed locally based on the calibration data to avoid privacy leakage risks and achieve safe and efficient health analysis and early warning processing.

[0130] Through the description of the above embodiments, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus the necessary general hardware platform, and of course can also be implemented by hardware, but in many cases the former is a better embodiment. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, can be embodied in the form of a software product, which is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk), and includes a number of instructions for enabling a terminal device (which can be a mobile phone, computer, server, air conditioner, or network device, etc.) to execute the methods described in each embodiment of the present invention.

[0131] The present invention can be used in a wide variety of general or special computer system environments or configurations. For example: personal computers, server computers, handheld or portable devices, tablet devices, multiprocessor systems, microprocessor-based systems, set-top boxes, programmable consumer electronics, network PCs, minicomputers, mainframe computers, distributed computing environments including any of the above systems or devices, and the like. The present invention can be described in the general context of computer-executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc. that perform specific tasks or implement specific abstract data types. The present invention can also be practiced in distributed computing environments in which tasks are performed by remote processing devices connected via a communications network. In a distributed computing environment, program modules can be located in local and remote computer storage media, including storage devices.

[0132] In summary, the present invention discloses a health data analysis method, apparatus, device and medium based on the Internet of Things, the method comprising: collecting multi-dimensional health sensor data and environmental data through multiple sensors in a local Internet of Things device; performing data fusion processing on the multi-dimensional health sensor data to obtain multimodal fusion sensor data; performing adaptive data calibration on the multimodal fusion sensor data according to the environmental data to obtain calibrated sensor data; inputting the calibrated sensor data into a dynamic early warning model pre-trained in the local Internet of Things device, and outputting corresponding health prompts after performing health dynamic early warning analysis on the calibrated sensor data locally. Data calibration of the multimodal fusion sensor data is performed through environmental data to improve the accuracy and reliability of health data analysis, and dynamic early warning analysis is performed locally based on the calibration data to avoid privacy leakage risks and achieve safe and efficient health analysis and early warning processing.

[0133] Of course, those skilled in the art will appreciate that all or part of the processes in the above-described method embodiments can be implemented by instructing related hardware (such as a processor, controller, etc.) through a computer program. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes in the above-described method embodiments. The storage medium can be a memory, a magnetic disk, a floppy disk, a flash memory, an optical storage device, etc.

[0134] It should be noted that if any software tools or components not developed by our company appear in the examples of this application, they are for illustration purposes only and do not represent actual use. It should be understood that the application of the present invention is not limited to the examples above. Those skilled in the art can make improvements or modifications based on the above description, and all such improvements and modifications shall fall within the scope of protection of the appended claims.

Claims

1. A health data analysis method based on the Internet of Things, characterized in that: include: Collect multi-dimensional health sensor data and environmental data through multiple sensors in local IoT devices; Performing data fusion processing on the multi-dimensional health sensor data to obtain multimodal fusion sensor data; Performing adaptive data calibration on the multimodal fusion sensor data according to the environmental data to obtain calibrated sensor data; The calibration sensor data is input into a dynamic early warning model pre-trained in the local Internet of Things device, and a corresponding health prompt is output after a health dynamic early warning analysis is performed on the calibration sensor data locally.

2. The health data analysis method based on the Internet of Things according to claim 1, characterized in that: The performing data fusion processing on the multi-dimensional health sensor data to obtain multimodal fusion sensor data includes: Performing noise reduction filtering and normalization processing on the multi-dimensional health sensor data to obtain multi-dimensional pre-processed sensor data; Obtaining correlations between pre-processed sensor data of different dimensions according to predefined data fusion rules, and determining fusion weights of different dimensions according to the correlations; The multi-dimensional pre-processed sensor data are weighted and summed according to the fusion weights of different dimensions to obtain multimodal fused sensor data.

3. The health data analysis method based on the Internet of Things according to claim 1, characterized in that: The adaptively calibrating the multimodal fusion sensor data according to the environmental data to obtain calibrated sensor data includes: Performing environmental feature analysis on the environmental data to extract environmental feature parameters related to health monitoring; confirming the user's motion state and body position state based on the multimodal fusion sensor data; Adaptively matching the environmental characteristic parameters, the user's motion state and body position state according to a preset calibration rule, and determining corresponding adaptive adjustment parameters; The multimodal fusion sensing data is calibrated according to the adaptive adjustment parameters to obtain the calibrated sensing data.

4. The method for analyzing health data based on the Internet of Things according to claim 1, characterized in that: The step of inputting the calibration sensor data into a dynamic early warning model pre-trained in the local IoT device, and outputting corresponding health prompts after performing a health dynamic early warning analysis on the calibration sensor data locally, includes: Inputting the calibrated sensor data into a pre-trained dynamic early warning model, wherein the dynamic early warning model is trained by federated learning on the local IoT device; Performing a health status analysis on the calibration sensor data locally using the dynamic early warning model to determine whether the current health indicator reaches the early warning threshold of the current user; If the warning threshold of the current user has been reached, a corresponding warning message is generated and a vibration and / or voice prompt is output through the local Internet of Things device.

5. The method for analyzing health data based on the Internet of Things according to claim 4, characterized in that: The dynamic warning model is obtained by performing federated learning training on the local IoT device through the following steps: Collecting basic health data and historical sensor data of the user, and constructing local training samples after preprocessing the basic health data and historical sensor data; Loading the initialized dynamic warning model, training the dynamic warning model on the local IoT device using the local training samples, updating model parameters and dynamically adjusting the user's warning threshold until a preset convergence condition is met; Uploading the local updated parameters of the dynamic warning model to the cloud server, so that the cloud server updates the global model according to the received multiple local updated parameters, and regularly distributes the latest global model parameters to all devices; The dynamic early warning model is optimized and updated according to the global model parameters received regularly.

6. The method for analyzing health data based on the Internet of Things according to claim 1, characterized in that: After inputting the calibration sensor data into the dynamic early warning model pre-trained in the local IoT device, and performing a health dynamic early warning analysis on the calibration sensor data locally and outputting a corresponding health prompt, the method further includes: When the health reminder includes warning information, identifying the warning event and the corresponding emergency level in the warning information; Confirm whether the emergency level of the warning event is higher than the preset level. If higher, output emergency response information to the family end, doctor end and / or hospital end bound to the local Internet of Things device according to the preset cascade response strategy.

7. The health data analysis method based on the Internet of Things according to claim 1, characterized in that: After inputting the calibration sensor data into the dynamic early warning model pre-trained in the local IoT device, and performing a local health dynamic early warning analysis on the calibration sensor data and outputting a corresponding health prompt, the method further includes: When the health reminder contains warning information, the corresponding encrypted metadata is generated according to the warning information and uploaded to the cloud server; When receiving a data sharing instruction from another device, the cloud server obtains corresponding zero-knowledge proof verification information from the cloud server and sends it to the requesting device, wherein the zero-knowledge proof verification information is obtained by the cloud server after performing zero-knowledge proof processing based on the encrypted metadata.

8. A health data analysis device based on the Internet of Things, characterized in that: include: A data acquisition module is used to collect multi-dimensional health sensor data and environmental data through multiple sensors in local IoT devices; A data fusion module is used to perform data fusion processing on the multi-dimensional health sensor data to obtain multimodal fusion sensor data; a data calibration module, configured to perform adaptive data calibration on the multimodal fusion sensor data according to the environmental data to obtain calibrated sensor data; The health dynamic analysis module is used to input the calibration sensor data into the dynamic warning model pre-trained in the local Internet of Things device, and output corresponding health prompts after performing health dynamic warning analysis on the calibration sensor data locally.

9. A computer device, characterized in that: comprising at least one processor; and, a memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the health data analysis method based on the Internet of Things according to any one of claims 1 to 7.

10. A non-volatile computer-readable storage medium, characterized in that: The non-volatile computer-readable storage medium stores computer-executable instructions, which, when executed by one or more processors, enable the one or more processors to execute the Internet of Things-based health data analysis method according to any one of claims 1 to 7.

Images