An internet-based health care remote monitoring method and system

Abstract

This invention relates to the field of nursing technology and provides an internet-based remote monitoring method and system for health care. The method includes: screening for deviation events and identifying risk monitoring nodes, extracting sampling data only for these risk monitoring nodes; performing comparative analysis on computational parameters in chronological order, comparing the magnitude of parameter fluctuations with preset fluctuation thresholds based on historical data statistics, and identifying key parameters with significant fluctuations; and generating predictive intervention nodes through a preset assessment model. This invention avoids indiscriminate processing of all data by locating effective data segments with high-risk characteristics from the raw data, reducing computational load and data transmission pressure, and improving the targeting and efficiency of risk assessment. By identifying fluctuation nodes where key parameters change, it can more sensitively capture the initial signs of potential risks and improve the accuracy of risk identification.

Description

Technical Field

[0001] This invention belongs to the field of nursing technology, specifically relating to an Internet-based remote health care monitoring method and system. Background Technology

[0002] With the development of the Internet of Things (IoT) and information technology, relying on IoT, communication, cloud computing and other Internet technologies, it is possible to collect, transmit and analyze patients' physiological data in real time outside the hospital and provide early warnings. At present, through continuous monitoring of patients' vital signs, it is possible to realize cross-regional allocation of medical resources and personalized health management, which can improve the level of chronic disease management, cope with population aging and improve patients' quality of life.

[0003] Existing remote health monitoring technologies struggle to distinguish between data changes caused by non-physiological factors such as environmental interference and poor contact, and risk signals genuinely arising from changes in a patient's condition. This can easily lead to false alarms and over-warnings, disrupting the normal workflow of healthcare workers and causing unnecessary psychological anxiety and burden on patients. When analyzing monitoring data, the focus is often solely on the amplitude of fluctuations caused by deviations from the normal range, neglecting key time-related characteristics such as the duration and frequency of fluctuations. In particular, short-lived and drastic fluctuations may simply be measurement artifacts, while small but persistent abnormalities may indicate potential health risks, affecting the timeliness of early intervention. Currently, most remote health monitoring technologies upload raw data to cloud servers for analysis. In scenarios with poor network conditions or requiring emergency response, data delays can occur, potentially delaying the optimal intervention time.

[0004] In view of this, this application provides a remote health care monitoring method and system based on the Internet. Summary of the Invention

[0005] The purpose of this invention is to provide a remote health care monitoring method and system based on the Internet to solve the technical problem in the prior art that the update status of body data cannot be identified, resulting in a large number of fluctuations in the updated data, which in turn leads to a significant increase in the fatigue of nursing staff.

[0006] The specific technical solution adopted by this invention is as follows: A remote health care monitoring method based on the Internet includes the following steps: Acquire historical operating data and real-time device parameters of user devices; Based on historical operational data, risk monitoring nodes are identified and key parameters are determined. Combined with the verification parameters generated based on real-time equipment parameters, intervention nodes are calculated and generated through a preset evaluation model. When it is determined that the parameters to be verified generated based on real-time device parameters meet the risk conditions defined by the intervention node, a risk alarm signal is output; The process involves identifying risk monitoring nodes and determining key parameters based on historical operational data, and combining these with verification parameters generated from real-time equipment parameters. An intervention node is then calculated using a pre-set evaluation model. This includes: screening for deviation events in equipment operation based on the user equipment's power-on duration and a pre-set power-on cycle to identify risk monitoring nodes; determining key parameters and associated fluctuation nodes based on sampling data corresponding to the risk monitoring nodes; and using the verification parameters calculated based on real-time equipment parameters, the pre-fluctuation nodes associated with the fluctuation nodes, and the post-fluctuation nodes as inputs to generate the intervention node through a pre-set evaluation model.

[0007] Preferably, the step of screening for deviation events in equipment operation based on the user equipment's power-on duration and preset power-on cycle to identify risk monitoring nodes includes: Determine whether the boot time is within the preset boot cycle time range. If the boot time is not within the time range, then a deviation event is determined to have occurred.

[0008] Preferably, the step of screening for deviation events in equipment operation based on the user equipment's power-on duration and a preset power-on cycle to identify risk monitoring nodes further includes: For deviation nodes where deviation events occur, the corresponding equipment status parameters are obtained and summarized as comparison data. The comparison data is analyzed, and if the difference between the values ​​of any two equipment status parameters is less than the fluctuation range threshold, or if the rate of change of the value of any equipment status parameter is greater than the rate of change threshold, then the deviation node corresponding to that equipment status parameter is identified as a risk monitoring node.

[0009] Preferably, determining the key parameters and the fluctuation nodes associated with the key parameters based on the sampling data corresponding to the risk monitoring nodes includes: The numerical deviation between the sampled data and the preset benchmark value is evaluated and processed to generate the calculation parameters corresponding to the numerical deviation. The numerical changes of the calculation parameters are compared and analyzed in chronological order to screen out the calculation parameters whose numerical fluctuation exceeds the preset fluctuation threshold, and these calculation parameters are identified as key parameters.

[0010] Preferably, the step of comparing and analyzing the numerical changes of the calculation parameters in chronological order, filtering out calculation parameters whose numerical fluctuations exceed a preset fluctuation threshold, and identifying these calculation parameters as key parameters includes: The preset fluctuation threshold is a boundary value calculated based on the statistical mean; if the value of the calculated parameter is higher than the boundary value of the 95% confidence interval calculated based on historical data, then the calculated parameter is determined as a key parameter.

[0011] Preferably, the step of comparing and analyzing the numerical changes of the calculation parameters in chronological order, filtering out calculation parameters whose numerical fluctuations exceed a preset fluctuation threshold, and identifying these calculation parameters as key parameters further includes: Calculate the time difference between the node before the fluctuation and the node after the fluctuation; if the time difference is less than the preset duration threshold, the key parameters are confirmed to be valid.

[0012] Preferably, the parameters to be verified generated based on the real-time device parameters include: Within two preset stable time intervals, calibration values ​​are obtained based on real-time device parameters. Preset correction operations are performed on the preset key reference parameters and the two obtained calibration values ​​to generate a first parameter to be checked and a second parameter to be checked. The first parameter to be checked is associated with the node before the fluctuation, and the second parameter to be checked is associated with the node after the fluctuation to form the parameter to be checked.

[0013] An internet-based remote health care monitoring system includes the following modules: The device status monitoring module is used to acquire historical operating data and real-time device parameters of user devices. The historical operating data includes at least the power-on duration and the preset power-on cycle. The intervention node generation module is used to identify risk monitoring nodes and determine key parameters based on historical operational data, and to calculate and generate intervention nodes by combining the parameters to be verified generated based on real-time equipment parameters and through a preset evaluation model. The risk alarm module is used to output a risk alarm signal when it is determined that the parameters to be verified generated based on real-time equipment parameters meet the risk conditions defined by the intervention node.

[0014] Preferably, risk monitoring nodes are identified and key parameters are determined based on historical operational data. Combined with verification parameters generated from real-time equipment parameters, intervention nodes are calculated and generated using a pre-set evaluation model, including: Based on the user equipment's power-on duration and preset power-on cycle, screening for deviation events in equipment operation is used to identify risk monitoring nodes. Based on the sampling data corresponding to the risk monitoring nodes, key parameters and fluctuation nodes associated with the key parameters are determined. The data of the parameters to be verified, the pre-fluctuation nodes and post-fluctuation nodes associated with the fluctuation nodes, generated by real-time equipment parameter correction calculations, are used as inputs and calculated through a preset evaluation model to generate intervention nodes.

[0015] Preferably, the step of screening for deviation events in equipment operation based on the user equipment's power-on duration and preset power-on cycle to identify risk monitoring nodes includes: Determine whether the boot time is within the preset boot cycle time range. If the boot time is not within the time range, then a deviation event is determined to have occurred. Beneficial effects

[0016] This invention screens deviation events based on power-on duration and preset power-on cycle, then analyzes the device status parameters corresponding to the deviation nodes to screen out risk monitoring nodes, and extracts sampling data only for the risk monitoring nodes. It locates effective data segments with high-risk characteristics from the original data, avoids indiscriminate processing of all data, reduces computational load and data transmission pressure, and improves the targeting and efficiency of risk assessment.

[0017] This invention compares and analyzes the computational parameters in chronological order, compares the magnitude of parameter fluctuations with preset fluctuation thresholds based on historical data statistics, filters out key parameters with significant fluctuations, and identifies the fluctuation nodes where key parameters change. It identifies the key parameters most sensitive to changes in health status from among many parameters, enabling more sensitive capture of the initial signs of potential risks and improving the accuracy of risk identification.

[0018] This invention utilizes data associated with identified pre-fluctuation and post-fluctuation nodes to calculate and generate predictive intervention nodes through a preset evaluation model. When making an alarm judgment, real-time equipment parameters are obtained to correct the benchmark key parameters, generating dynamic parameters to be checked. These parameters are then compared with the intervention nodes to improve prediction accuracy. By dynamically checking the judgment criteria with real-time parameters, measurement deviations introduced by non-physiological factors such as changes in equipment status are compensated, enabling alarm judgments to match the user's immediate state, reducing the probability of false alarms and missed alarms, and ensuring the reliability of remote monitoring. Attached Figure Description

[0019] Figure 1 This is a flowchart of the method of the present invention; Figure 2 This is a system module diagram of the present invention. Detailed Implementation

[0020] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to specific embodiments. It should be understood that the specific embodiments described herein are merely for explaining the invention and are not intended to limit the scope of protection of the invention. Example

[0021] Please refer to Figure 1 This embodiment provides a remote health care monitoring method based on the Internet, including the following steps: S1. Obtain basic information about the user equipment. Basic information consists of basic parameters that reflect the physical attributes of the equipment itself and the preset operating plan. Obtaining basic information about the user equipment is used to provide a benchmark for subsequent analysis of the stability of user habits. Among them, user equipment is a terminal device that can continuously or periodically collect users' vital signs or activity data. User equipment includes devices such as electrocardiogram monitors, pulse oximeters, smart bracelets, smart mattresses, or continuous glucose monitors. Basic information includes device brand, model, power-on duration, and preset power-on cycle. Among them, device brand and model are used to associate specific hardware performance baselines and data accuracy, while power-on duration and preset power-on cycle are directly used for preliminary risk screening. The preset power-on cycle is a specific daily time range that the user device is expected to be powered on, which is used to compare with the actual power-on duration to identify deviations in user habits. When a user device is first activated, basic information is entered through the built-in information collection module of the user device, and a unique binding relationship is established with the user account on the server. After the binding is completed, the user device can maintain real-time or timed data interaction with the server through Internet communication protocols such as MQTT or HTTP, providing a data foundation for continuous monitoring and analysis.

[0022] S2. Preliminarily screen out time points with potential risks from the data; The data is initially screened to identify time points with potential risks, specifically including: screening for deviation events, based on the obtained power-on duration and preset power-on cycle, screening for deviation events in equipment operation; Deviation events in the operation of screening equipment include determining whether the power-on duration is within the time range limited by the preset power-on cycle; specifically, for health monitoring equipment set for a specific user, the preset power-on cycle can be set to a specific time period every day, such as a sleep monitor which can set the preset power-on cycle to 8 pm to 8 am the next day. If the device's power-on time on a certain day exceeds the start or end time of the time range, it is determined that the user's usage habits have changed. This change may be related to abnormalities in health status or lifestyle. The deviation event is identified, and the time of the deviation event is recorded as a deviation node. For each deviation node, the device status parameters at the corresponding time are obtained. Among them, the equipment status parameters are a set of data that reflect the physical working environment and working conditions of the equipment, including the internal temperature of the equipment, the remaining battery power, the sensor connection status or wireless signal strength, etc. The core function of the equipment status parameters is to provide a basis for distinguishing between user physiological abnormalities and equipment malfunctions. All acquired equipment status parameters are summarized into comparison data. Based on the comparison data, risk monitoring nodes that meet the preset risk conditions are selected. From the deviation nodes with abnormal usage habits, nodes that may cause the monitoring data to be distorted due to abnormal equipment or drastic changes in the external environment are identified. The preset risk conditions include: determining whether the value of any device status parameter in the comparison data remains within a very small fluctuation range within a preset period of time, that is, whether the standard deviation or range of its value change is less than the fluctuation range threshold, and identifying faults such as sensor malfunction or data transmission freeze. If this condition is met, the corresponding deviation node is identified as a risk monitoring node. Specifically, when the user equipment is a temperature sensor, its fluctuation range threshold can be set to less than 0.1℃. When the temperature sensor outputs the same value for a long time or its standard deviation or range of value change is less than or continuously less than 0.1℃, it can be considered that the temperature sensor has a fault such as malfunction or data transmission freeze, and the corresponding deviation node is identified as a risk monitoring node. The preset risk conditions may also include: determining whether the rate of change of any device status parameter in the comparison data is greater than the rate of change threshold, and identifying sudden changes in the device's working environment, such as a sudden temperature rise due to being covered; if this condition is met, the corresponding deviation node is also identified as a risk monitoring node; specifically, when the user device is a temperature sensor, its rate of change threshold can be set to 2.0℃ / min. When the rate of change of the device status parameter of the temperature sensor is greater than 2.0℃ / min, a sudden change in the working environment of the temperature sensor is identified, and the corresponding deviation node is identified as a risk monitoring node. Monitoring data is acquired from user devices, and the monitoring data within the time period corresponding to the risk monitoring node is extracted as sampling data. Through screening, the sampling data is limited to the time window with the most analytical value, improving the efficiency and targeting of subsequent processing. Among them, the monitoring data is physiological data that directly reflects the user's vital signs, including heart rate, blood oxygen saturation, respiratory rate or body temperature.

[0023] S3. Transform the raw and discrete sampled data into structured and easily quantifiable operational parameters; perform evaluation processing based on the numerical deviation between the sampled data and the preset benchmark value, and generate operational parameters corresponding to the numerical deviation. To ensure the traceability and integrity of the evaluation process, for each data point in the sampled data, a calculation parameter consisting of multiple parameter items is generated. The calculation parameter includes at least three parameter items: the original sampled data value, the difference calculation result calculated based on the sampled data and the preset benchmark value, and the calculation operation expression used to record the difference calculation process. Specifically, if the sampled data at a certain moment shows a blood oxygen saturation value of 92%, while the preset benchmark value is 97%, then the corresponding difference calculation result is -5%. The calculation operation expression is recorded as a text string, such as blood oxygen saturation value 92% - preset benchmark value 97%. The calculation operation expression records how this difference is calculated. Thus, the blood oxygen saturation value of 92%, the difference calculation result of -5%, and the blood oxygen saturation value of 92% - preset benchmark value of 97% are collectively encapsulated as calculation parameters. This allows subsequent analysis to not only obtain the difference calculation results, but also to instantly trace the results back to the original reading of blood oxygen saturation at 92% and the comparison calculation with the preset benchmark value of 97%. The self-interpretive data structure provides the necessary contextual information for diagnostic logic and debugging.

[0024] S4. Identify the most indicative dramatic changes from the time series fluctuations of a series of operational parameters; compare and analyze the generated difference calculation results in chronological order, screen out operational parameters whose numerical fluctuation amplitude exceeds the preset fluctuation threshold, and determine the operational parameter as the key parameter. The preset fluctuation threshold is a dynamically calculated boundary value, rather than a fixed constant. It is calculated based on the user's set of calculation parameters over a recent historical period, which can be the past seven days, using statistical methods to determine the upper or lower boundary of a 95% confidence interval. This allows the threshold to automatically adapt to the normal fluctuation range exhibited by a specific user over a long period of time, thereby effectively filtering out common statistical noise and ensuring that the selected fluctuations are statistically significant. If the value of a certain operation parameter exceeds the limit value, it is considered an anomaly event worthy of attention, and the operation parameter is identified as a critical parameter; after identifying the critical parameter, the moment associated with the critical parameter is identified as the fluctuation node; To further confirm the validity of the fluctuation and avoid misjudging isolated and transient pulse interference as meaningful events, a verification procedure was introduced. The verification procedure includes: setting the time node immediately before the fluctuation node as the pre-fluctuation node and the time node immediately after the fluctuation node as the post-fluctuation node; calculating the time difference between the pre-fluctuation node and the post-fluctuation node, and comparing the time difference with a preset duration threshold, which can be 5 minutes. If the time difference is less than the preset duration threshold, it indicates that the change is a drastic change that occurs in a short period of time, which is consistent with the characteristics of an acute risk event. Therefore, the key parameter is confirmed to be valid. If the time difference is not less than the preset duration threshold, it indicates that the change is a slow long-term trend or isolated noise that has been smoothed out by subsequent data. It does not have the value of immediate intervention. Therefore, the key parameter is determined to be invalid and is removed from the subsequent analysis process.

[0025] S5. Dynamically calculate personalized risk thresholds as intervention nodes, thereby freeing alarm judgment from dependence on universal fixed thresholds; The calculation of the intervention node includes: obtaining the real-time equipment parameters of the user equipment, and at the same time obtaining the preset benchmark key parameters. The benchmark key parameters are theoretically standardized parameter values, which can be the normal values ​​of a certain indicator defined in a physiology textbook; performing preset correction calculations on the benchmark key parameters based on the real-time equipment parameters to generate parameters to be checked, and performing personalized corrections on the standardized benchmark values ​​according to the current actual operating conditions of the equipment. The parameters to be verified include: identifying two relatively stable time intervals for physiological indicators in the user's recent historical data, the stable time intervals can be the deep sleep stages of the user on two different nights; within the two preset stable time intervals, obtaining calibration values ​​that can reflect the device's operating conditions, the calibration values ​​can be the average ambient temperature or the average battery voltage at that time. The benchmark key parameters and the two obtained calibration values ​​are respectively subjected to preset correction calculations to generate the first parameter to be checked and the second parameter to be checked. The preset correction calculation can follow the weighted correction procedure. The preset correction calculation is: parameter to be checked = benchmark key parameter × (1 + weight coefficient × (current calibration value - calibration baseline value)). The weighting coefficient determines the degree of influence of the deviation between the current calibration value and its baseline value on the final parameter to be calibrated. Its value can be determined empirically based on the specific equipment model and environmental factors. The calibration baseline value is the standard reference value of the parameter used for calibration, such as ambient temperature or battery voltage. Specifically, when the equipment parameter is temperature, the preset benchmark key parameter can be -3.0%, the calibration baseline value is 30°C, and the weighting coefficient is 0.05. The first parameter to be verified is associated with the determined node before the fluctuation, and the second parameter to be verified is associated with the node after the fluctuation, thus forming the fitting verification data. The fitting verification data essentially constructs two anchor points; among them, the two anchor points define a personalized safety baseline based on the user's real stable state and equipment operating conditions. Using the fitted verification data as input, the system calculates and generates intervention nodes through a preset evaluation model. Specifically, the preset evaluation model treats the node before fluctuation and the first parameter to be verified, and the node after fluctuation and the second parameter to be verified contained in the fitted verification data as two coordinate points with a linear relationship. The calculation relationship of the numerical change trend is established through these two coordinate points, and the critical value is calculated based on this relationship. This critical point is the intervention node, which represents the level to which the key parameters of a specific user should be considered as a risk state requiring intervention under the current equipment operating conditions.

[0026] S6. Runs continuously in the background, acquires real-time device parameters during device operation, and generates risk alarm signals. Based on the currently acquired real-time device parameters, a preset correction calculation is performed on the benchmark key parameters to generate real-time updated parameters to be checked for alarm judgment. The values ​​of the parameters to be checked are compared with the values ​​of the generated intervention nodes in real time. The preset correction calculation uses the exact same calculation as the calculation of the parameters to be checked. If the value of the parameter to be checked used for alarm judgment is equal to or greater than the value of the intervention node, it means that the user's real-time health indicators have reached the risk threshold tailored to them, and a risk alarm signal is immediately output. The risk alarm signal can take many forms, such as sending a high-priority notification to the user's terminal device, or sending a text message or automatic voice call to a preset family member or nursing center. The risk alarm signal can realize hierarchical alarm. If the value of the parameter to be checked is equal to that of the intervention node, a lower-level warning signal such as a yellow indicator will be displayed on the monitoring interface. If the value is greater than that of the intervention node, the highest-level risk alarm signal such as a red indicator and an audible alarm will be displayed on the output interface. If the value of the parameter to be checked used for alarm judgment is always lower than that of the intervention node, the silent monitoring state will be maintained to avoid unnecessary interference. Example

[0027] Please refer to Figure 2 This embodiment provides an Internet-based remote health monitoring system. In its implementation, this system can be deployed on a cloud server, a private server, or a dedicated health monitoring center computing platform. It securely communicates with various user devices deployed on the user's end via the Internet, such as portable ECG monitors, smart blood pressure monitors, or blood glucose meters. The system includes the following modules: The device status monitoring module is used to acquire and manage data from user devices; it receives and stores historical operating data and real-time device parameters from user devices connected to the Internet. Historical operation data includes at least the records of the user device’s power-on duration each time, as well as the preset power-on cycle for the device. The preset power-on cycle can be a time range, specifically defining the minimum and maximum duration that the device should run continuously after each normal power-on. Real-time device parameters are the physiological indicators or other status readings that the user device is currently measuring, providing a complete and orderly data foundation for subsequent analysis.

[0028] The intervention node generation module is used to perform analysis and modeling tasks and generate intervention nodes for risk assessment. Based on historical operating data provided by the equipment status monitoring module, deviation events in equipment operation are screened to identify risk monitoring nodes; it is determined whether the power-on duration of a user's equipment operation is within the time range limited by the preset power-on cycle. If the power-on time is too short or too long, causing the power-on duration to be outside the preset power-on cycle, then a deviation event is determined to have occurred, and the record corresponding to that time point is marked as a deviation node. For deviation nodes where deviation events occur, multiple device status parameters corresponding to the deviation node are acquired and summarized into comparison data. The device status parameters may include the battery voltage or sensor baseline value during device self-test. By analyzing the comparison data, if the difference between the values ​​of any two device status parameters is less than the fluctuation range threshold, it indicates that the device sensor or a certain component is stuck or unresponsive. Alternatively, if the rate of change of the value of any device status parameter is greater than a rate of change threshold, it indicates that the device has experienced an abnormal instantaneous impact or interference. In this case, the deviation node corresponding to the device status parameter is finally determined as a risk monitoring node. After identifying the risk monitoring node, the key parameters and the fluctuation nodes associated with the key parameters are determined based on the sampling data corresponding to the risk monitoring node, which is the physiological measurement data within a period of time before and after the node. The numerical deviation between the sampled data and the preset benchmark value is evaluated and processed to generate the calculation parameters corresponding to the numerical deviation. The numerical changes of the calculation parameters are compared and analyzed in chronological order to screen out the calculation parameters whose numerical fluctuation exceeds the preset fluctuation threshold. The preset fluctuation threshold can be a boundary value calculated based on the statistical average of historical data. If the value of a certain operation parameter is higher than the upper boundary value of the 95% confidence interval calculated based on a large amount of historical data, it is considered that its fluctuation range is significant. The selected operation parameter is determined as the key parameter of this event, and the moment when it fluctuates sharply is recorded as the fluctuation node. At the same time, the pre-fluctuation node immediately before it and the post-fluctuation node immediately after it are also recorded. To ensure the validity of the identified key parameters, the time difference between the node before the fluctuation and the node after the fluctuation is calculated. If the time difference is less than the preset duration threshold, the key parameter is confirmed to be valid, which helps to eliminate slow and long-term drift and focus on sudden changes. Based on the real-time device parameter correction calculation, the parameters to be verified are generated. Two preset stable time intervals are found in the real-time device parameter stream, and calibration values ​​are obtained based on the real-time device parameters within these two intervals. The benchmark key parameter and the two obtained calibration values ​​are respectively subjected to preset correction calculations. The benchmark key parameter is the standard value of the key parameter under ideal conditions. The first parameter to be verified and the second parameter to be verified are generated. The first parameter to be verified is associated with the node before the fluctuation in the historical event, and the second parameter to be verified is associated with the node after the fluctuation in the historical event, together forming the fitting verification data. The fitting and verification data are used as input, and calculations are performed through a preset evaluation model. The preset evaluation model can be a trained machine learning classifier, a complex rule engine, or a multi-dimensional vector space similarity matching model. The preset evaluation model generates intervention nodes by comprehensively analyzing the similarity and correlation between real-time data patterns and historical risk event patterns. The intervention node encapsulates specific risk conditions, including risk level scores, Boolean risk indicators, or specific parameter thresholds that trigger alarms.

[0029] The risk alarm module is used to perform the final risk assessment and response. It continuously receives intervention nodes generated by the intervention node generation module and compares the verification parameters for alarm assessment generated based on real-time device parameters with the risk conditions defined by the intervention nodes in real time. When the matching degree between the real-time parameter mode and the risk mode defined by the intervention node exceeds the threshold, and it is determined that the parameter to be checked meets the risk conditions, a risk alarm signal is immediately output. This risk alarm signal can be pushed to the monitoring platform of medical staff, sent to designated family members or guardians, or directly reminded to the user in the form of sound and light on the user's device, thereby realizing timely early warning of potential health risks.

[0030] The above embodiments are only for illustrating the technical concept and features of the present invention, and are intended to enable those skilled in the art to understand and implement the present invention. They should not be construed as limiting the scope of protection of the present invention. All equivalent changes or modifications made in accordance with the spirit and essence of the present invention should be covered within the scope of protection of the present invention.

Claims

1. A remote health care monitoring method based on the Internet, characterized in that, Includes the following steps: Acquire historical operating data and real-time device parameters of user devices; Based on historical operational data, risk monitoring nodes are identified and key parameters are determined. Combined with the verification parameters generated based on real-time equipment parameters, intervention nodes are calculated and generated through a preset evaluation model. When it is determined that the parameters to be verified generated based on real-time device parameters meet the risk conditions defined by the intervention node, a risk alarm signal is output; The process involves identifying risk monitoring nodes and determining key parameters based on historical operational data, and combining these with verification parameters generated from real-time equipment parameters. An intervention node is then calculated using a pre-set evaluation model. This includes: screening for deviation events in equipment operation based on the user equipment's power-on duration and a pre-set power-on cycle to identify risk monitoring nodes; determining key parameters and associated fluctuation nodes based on sampling data corresponding to the risk monitoring nodes; and using the verification parameters calculated based on real-time equipment parameters, the pre-fluctuation nodes associated with the fluctuation nodes, and the post-fluctuation nodes as inputs to generate the intervention node through a pre-set evaluation model.

2. The method for remote health care monitoring based on the Internet according to claim 1, characterized in that, The process of screening for deviation events in equipment operation based on the user equipment's power-on duration and preset power-on cycle to identify risk monitoring nodes includes: Determine whether the boot-up time is within the preset boot-up cycle time range. If the boot-up time is not within the time range, then a deviation event is determined to have occurred.

3. The method for remote health care monitoring based on the Internet according to claim 2, characterized in that, The method of screening for deviation events in equipment operation based on the user equipment's power-on duration and preset power-on cycle to identify risk monitoring nodes also includes: For deviation nodes where deviation events occur, the corresponding equipment status parameters are obtained and summarized as comparison data. The comparison data is analyzed, and if the difference between the values ​​of any two equipment status parameters is less than the fluctuation range threshold, or if the rate of change of the value of any equipment status parameter is greater than the rate of change threshold, then the deviation node corresponding to that equipment status parameter is identified as a risk monitoring node.

4. The Internet-based remote health care monitoring method according to claim 1, characterized in that, The process of determining key parameters and associated fluctuation nodes based on sampling data corresponding to risk monitoring nodes includes: The numerical deviation between the sampled data and the preset benchmark value is evaluated and processed to generate the calculation parameters corresponding to the numerical deviation. The numerical changes of the calculation parameters are compared and analyzed in chronological order to screen out the calculation parameters whose numerical fluctuation exceeds the preset fluctuation threshold, and these calculation parameters are identified as key parameters.

5. The Internet-based remote health care monitoring method according to claim 4, characterized in that, The step of comparing and analyzing the numerical changes of the calculation parameters in chronological order, filtering out calculation parameters whose numerical fluctuations exceed a preset fluctuation threshold, and identifying these calculation parameters as key parameters includes: The preset fluctuation threshold is a boundary value calculated based on the statistical mean; if the value of the calculated parameter is higher than the boundary value of the 95% confidence interval calculated based on historical data, then the calculated parameter is determined as a key parameter.

6. The Internet-based remote health care monitoring method according to claim 5, characterized in that, The step of comparing and analyzing the numerical changes of the calculation parameters in chronological order, filtering out calculation parameters whose numerical fluctuations exceed a preset fluctuation threshold, and identifying these calculation parameters as key parameters also includes: Calculate the time difference between the node before the fluctuation and the node after the fluctuation; if the time difference is less than the preset duration threshold, the key parameters are confirmed to be valid.

7. The Internet-based remote health care monitoring method according to claim 1, characterized in that, The parameters to be verified, generated based on real-time device parameters, include: Within two preset stable time intervals, calibration values ​​are obtained based on real-time device parameters. Preset correction operations are performed on the preset key reference parameters and the two obtained calibration values ​​to generate a first parameter to be checked and a second parameter to be checked. The first parameter to be checked is associated with the node before the fluctuation, and the second parameter to be checked is associated with the node after the fluctuation to form the parameter to be checked.

8. A remote health care monitoring system based on the Internet, characterized in that, Includes the following modules: The device status monitoring module is used to acquire historical operating data and real-time device parameters of user devices. The historical operating data includes at least the power-on duration and the preset power-on cycle. The intervention node generation module is used to identify risk monitoring nodes and determine key parameters based on historical operational data, and to calculate and generate intervention nodes by combining the parameters to be verified generated based on real-time equipment parameters and through a preset evaluation model. The risk alarm module is used to output a risk alarm signal when it is determined that the parameters to be verified generated based on real-time equipment parameters meet the risk conditions defined by the intervention node.

9. A remote health care monitoring system based on the Internet according to claim 8, characterized in that, Based on historical operational data, risk monitoring nodes are identified and key parameters are determined. Combined with verification parameters generated from real-time equipment parameters, intervention nodes are calculated and generated using a pre-set evaluation model. Based on the user equipment's power-on duration and preset power-on cycle, screening for deviation events in equipment operation is used to identify risk monitoring nodes. Based on the sampling data corresponding to the risk monitoring nodes, key parameters and fluctuation nodes associated with the key parameters are determined. The data of the parameters to be verified, the pre-fluctuation nodes and post-fluctuation nodes associated with the fluctuation nodes, generated by real-time equipment parameter correction calculations, are used as inputs and calculated through a preset evaluation model to generate intervention nodes.

10. A remote health care monitoring system based on the Internet according to claim 9, characterized in that, The process of screening for deviation events in equipment operation based on the user equipment's power-on duration and preset power-on cycle to identify risk monitoring nodes includes: Determine whether the boot-up time is within the preset boot-up cycle time range. If the boot-up time is not within the time range, then a deviation event is determined to have occurred.

Images