Remote home care detection system for old people in community

Abstract

The invention relates to the technical field of remote home care detection. The invention relates to a remote home care detection system for old people in a community. The system comprises a connection establishment unit, an old people risk analysis unit, a body fluctuation analysis unit, a body abnormity marking unit and a rescue call management unit. The connection establishment unit is used for connecting the household intelligent equipment and a community medical center and performing correlation analysis on the body data type collected by the household intelligent equipment in combination with brain diseases; seamless connection between equipment and a community medical center is realized through an intelligent management module, the community medical center is quickly responded to treat and cure the old people, and meanwhile, an acquisition reminding module guides the old people to complete supplementary acquisition based on data association degree and weight sorting and simple interaction, so that the operation threshold is reduced, and the operation efficiency is improved. And if the elderly cannot complete acquisition or the threshold value is still exceeded after supplementary acquisition, an emergency notification and position information are automatically sent to form monitoring-reminding-rescue.

Description

Technical Field

[0001] The present invention relates to the technical field of remote home care detection, in particular to a remote home care detection system for elderly people in a community. Background Art

[0002] In the field of community-based elderly care, as the aging of the population deepens, remote health monitoring technology for the elderly has become an important support for ensuring the safety of home-based elderly care. Existing technologies mainly collect the physical data of the elderly through smart wearable devices, health monitoring instruments, etc., and transmit them to community medical centers or family members through the network, realizing real-time monitoring of the basic health status of the elderly and abnormal warning.

[0003] However, existing technologies mainly focus on the monitoring of single body data, and application scenarios are mostly limited to the collection and transmission of basic vital signs. There is a lack of comprehensive analysis of multi-dimensional data such as the elderly's behavioral patterns and environmental conditions. When the elderly fall or work at home, the body data fluctuates, which may lead to the misconception that brain disease has occurred. Therefore, relying solely on physiological indicators may not be able to timely identify potential risks such as falls caused by mobility problems and long-term inactivity. It is easy to cause misjudgment or missed judgment due to the single data dimension. In order to reduce this situation, a remote home care detection system for community elderly people is proposed. Summary of the Invention

[0004] The purpose of the present invention is to provide a remote home care detection system for community elderly people to solve the problems raised in the above background technology.

[0005] To achieve the above-mentioned purpose, a remote home care detection system for the elderly in the community is provided, which includes a connection establishment unit, an elderly risk analysis unit, a body fluctuation analysis unit, a body abnormality marking unit, and a rescue call management unit; The connection establishment unit is used to connect the home smart device and the community medical center, and perform correlation analysis on the body data type collected by the home smart device in combination with brain diseases; The elderly risk analysis unit is used to obtain physical information of the elderly, determine physical data related to brain diseases based on the physical information, and set a risk status for the physical data based on the physical information; The body fluctuation analysis unit is used to collect family image data, perform behavior analysis on the elderly based on the family image data, determine the elderly's physical condition based on the behavior analysis results, and perform normal fluctuation trend analysis based on the physical condition combined with historical physical data; The body abnormality marking unit is used to perform abnormal comparison on each type of real-time body data in combination with the risk status and the normal fluctuation trend, and mark the real-time body data of that type as abnormal according to the comparison result; The rescue call management unit is used to assign brain disease weights and abnormal thresholds to each type of physical data in combination with physical information, and then compare the weights of the abnormally marked real-time physical data with the abnormal thresholds, and select the physical data type with the highest correlation and not collected when the real-time physical data is abnormal to remind the elderly to collect it.

[0006] As a further improvement of the present technical solution, the connection establishment unit establishes an intelligent management module through a home network terminal, and then establishes a connection between the intelligent management module and the home smart device, thereby obtaining the physical data collected by the home smart device on the elderly, and at the same time establishes a connection between the intelligent management module and the community medical center, thereby sending emergency notifications and home locations to the community medical center, so that the community medical center can provide assistance to the elderly.

[0007] As a further improvement of the present technical solution, the connection establishment unit includes a correlation analysis module; The correlation analysis module is used to identify the type of body data collected by home smart devices, obtain the data type corresponding to each body data, and then obtain the data type associated with each brain disease through medical atlas, so as to perform brain disease-related matching on each body data type of home smart devices.

[0008] As a further improvement of this technical solution, the elderly risk analysis unit includes a body information acquisition module and a risk setting module; The body information acquisition module is used to collect body information related to the elderly from the home manager, identify brain diseases based on the relevant body information, and obtain the corresponding brain diseases and basic body information of the elderly based on the identification results; The risk setting module is used to set the risk status according to the brain diseases and basic physical information corresponding to the elderly combined with the physical data collected by home smart devices, and set the corresponding risk status for each type of physical data.

[0009] As a further improvement of the present technical solution, the body fluctuation analysis unit includes an image acquisition module and a trend analysis module; The image acquisition module is used to collect indoor image data through home smart devices and aggregate the collected indoor image data into home image data; The trend analysis module is used to perform behavioral analysis on the elderly based on family image data, obtain the behavior type the elderly are currently in, and then perform physical condition analysis based on the behavior type combined with basic physical information to obtain the elderly's current physical condition, and then perform normal fluctuation trend analysis on the physical condition combined with historical physical data to obtain the normal fluctuation trend corresponding to each type of physical data.

[0010] As a further improvement of the present technical solution, the body fluctuation analysis unit obtains a collection timestamp of the body data and divides the body data into real-time body data and historical body data according to the collection timestamp; The collection timestamps adjacent to the real time are regarded as real-time body data, and the others are regarded as historical body data.

[0011] As a further improvement of the present technical solution, the body abnormality marking unit includes an abnormality comparison module and a marking summary module; The abnormality comparison module is used to perform real-time fluctuation analysis on each type of real-time body data, compare the real-time fluctuation with the normal fluctuation trend, and mark it as abnormal when the real-time fluctuation exceeds the normal fluctuation trend. Otherwise, the real-time fluctuation does not exceed the normal fluctuation trend, and the monitoring is continued; Perform predictive analysis of body data based on real-time fluctuations combined with real-time body data, and compare the predicted body data with the risk status. If the predicted body data is in a risky state, it will be marked as abnormal. Conversely, if the predicted body data is not in a risky state, it will continue to be monitored. The marking summary module is used to perform abnormal summary statistics on the marked real-time body data.

[0012] As a further improvement of the present technical solution, the rescue call management unit includes an allocation setting module, a weight comparison module, and a collection reminder module; The allocation setting module is used to allocate brain disease weights to each type of physical data in combination with basic physical information, obtain the weight of each type of physical data for the onset of brain disease, and set abnormal thresholds according to the corresponding brain disease and basic physical information; The weight comparison module is used to compare the abnormal real-time body data of the abnormal summary statistics with the corresponding weight and the abnormal threshold. When the weight occupied by the abnormal real-time body data statistics is greater than the abnormal threshold, the community medical center is called for rescue and examination. On the contrary, when the weight occupied by the abnormal real-time body data statistics is less than the abnormal threshold, the collection reminder module is entered; The collection reminder module is used to obtain the body data type that has not been collected in real time based on the real-time body data type, and at the same time, according to the difference between the weight of the collected quantity and the abnormal threshold, select the uncollected body type associated with the abnormal actual review body data and sort it according to the proportion weight, and remind the elderly to go to the home smart device corresponding to the highest-ranked uncollected body type to collect body data. When the elderly are unable to complete the collection, call the community medical center for rescue and examination. After the elderly complete the body data collection, the weight is updated and compared with the abnormal threshold for a second time.

[0013] As a further improvement of the present technical solution, the number of collected data in the collection reminder module is the total number of body types corresponding to the real-time body data, and the number of uncollected data represents the total number of body types that can be collected by the home smart device excluding the body types corresponding to the number of collected data.

[0014] Compared with the prior art, the present invention has the following beneficial effects: 1. This is a remote home care detection system for the elderly in the community. It realizes seamless connection between the equipment and the community medical center through the intelligent management module, and quickly responds to the community medical center to treat the elderly. At the same time, the collection reminder module is based on data correlation and weight sorting, and simple interaction is used to guide the elderly to complete the re-collection, thereby lowering the operation threshold. If the elderly cannot complete the collection or the threshold is still exceeded after the re-collection, emergency notification and location information will be automatically sent to form a monitoring-reminder-rescue system. In addition, the application of medical maps and machine learning models realizes dynamic optimization of risk assessment, can adapt to individual differences, continuously improve service reliability, and provide safe and convenient remote care solutions for the elderly in the community.

[0015] 2. This system is used in remote home care monitoring systems for the elderly in the community. It uses smart devices to synchronously collect physiological data and behavioral data to build a dual monitoring system. By comparing real-time blood pressure fluctuations and analyzing the frequency of abnormal gait, it can accurately identify potential risks of stroke and avoid misjudgment of single data. It fills the gap of traditional monitoring that relies only on physiological indicators and improves the comprehensiveness and accuracy of health assessment.

[0016] 3. This is a remote home care detection system for the elderly in the community. Based on basic information such as the elderly's medical history and age, differentiated weights are assigned to different data types through medical maps, and dynamic risk thresholds are set. The dual anomaly detection mechanism ensures timely risk warnings. At the emergency response level, the weight comparison module directly initiates rescue based on the total weight of the abnormal data. When the weight does not meet the standard, the collection reminder module guides the collection of related data, forming an efficient mechanism for accurate warning and graded response, and optimizing the allocation of medical resources. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 It is the overall structural principle diagram of the present invention.

[0018] The meaning of each number in the figure is: 10. Connection establishment unit; 20. Elderly risk analysis unit; 30. Body fluctuation analysis unit; 40. Body abnormality marking unit; 50. Rescue call management unit. DETAILED DESCRIPTION

[0019] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0020] See also Figure 1 As shown, the purpose of this embodiment is to provide a remote home care detection system for community elderly people, including a connection establishment unit 10, an elderly risk analysis unit 20, a body fluctuation analysis unit 30, a body abnormality marking unit 40 and a rescue call management unit 50; The connection establishment unit 10 is used to connect the home smart device and the community medical center, and perform correlation analysis on the body data collected by the home smart device in combination with brain diseases; The connection establishment unit 10 establishes an intelligent management module through the home network terminal, and then establishes a connection between the intelligent management module and the home smart device to obtain the physical data collected by the home smart device on the elderly. At the same time, the intelligent management module is connected to the community medical center to send an emergency notification and the home location to the community medical center, so that the community medical center can provide assistance to the elderly. The specific steps are as follows: Intelligent management module construction: Through the home network terminal, set the network parameters of the intelligent management module, initialize the software of the intelligent management module, load the relevant drivers and management system, and enable it to have the ability to operate and interact; Connecting with smart home devices: The intelligent management module scans surrounding smart home devices, such as smart bracelets and smart blood pressure monitors, through wireless communication protocols such as Bluetooth and Wi-Fi. After identifying the device, it performs identity authentication and pairing operations according to the device's pairing rules, establishing a stable data transmission channel, thereby obtaining the elderly's physical data collected by the smart home devices, such as heart rate, blood pressure, and step count; Connecting with the community medical center: The intelligent management module establishes a network connection with the server of the community medical center through the Internet, using secure network protocols such as HTTPS; Send emergency notifications and rescue responses: The intelligent management module continuously analyzes the physical data of the elderly. When the data is abnormal, it sends an emergency notification to the community medical center, and also attaches the home location information obtained through the positioning function; After receiving the notification, the community medical center arranges medical staff to go to the elderly’s residence to provide assistance based on the location information.

[0021] The connection establishment unit 10 includes a correlation analysis module; The correlation analysis module is used to identify the type of physical data collected by home smart devices, obtain the data type corresponding to each physical data, and then obtain the data type associated with each brain disease through medical atlas, thereby performing brain disease-related matching for each physical data type of home smart devices. The specific steps are as follows: Data type identification: Remove noise, missing values, and outliers from the collected data, extract key features from the raw data, and then map the extracted features to standard data types (blood pressure, heart rate, blood sugar, etc.) through predefined rules or machine learning models; Medical graph construction and query: Associate brain diseases (stroke, Alzheimer's disease) with related data types (blood pressure, cognitive scores) to build a medical graph (graph structure). Query the set of associated data types corresponding to each brain disease using a graph traversal algorithm (breadth-first search); Data type and disease matching: The identified data type is compared with the disease-related data type in the medical atlas, and a weight is assigned to each match based on the strength of the association (for example, the association weight of hypertension to stroke is higher than that of body temperature).

[0022] The elderly risk analysis unit 20 is used to obtain physical information of the elderly, determine physical data related to brain diseases based on the physical information, and set a risk status for the physical data based on the physical information; The elderly risk analysis unit 20 includes a body information acquisition module and a risk setting module; The physical information acquisition module is used to collect physical information related to the elderly from the home manager, identify brain diseases based on the relevant physical information, and obtain the corresponding brain diseases and basic physical information of the elderly based on the identification results; Through questionnaires, we collect information such as the elderly's medical history, allergy history, medication usage, daily activity ability, etc. from home managers, and organize the collected unstructured text information into standardized fields, such as disease name, onset time, severity, etc.

[0023] Determination of brain disorders in the elderly is based on information provided by furniture managers.

[0024] The risk setting module is used to set risk status based on the elderly's corresponding brain diseases and basic physical information combined with the physical data collected by home smart devices. The corresponding risk status is set for each type of physical data. The specific formula is as follows: ; Among them, R base is the basic risk score, n is the number of body data, w i is the weight of the i-th body data, p i is the normalized value of the i-th body data; ; ; Among them, T low and T high are the lower and upper limits of the risk threshold, μ and σ are the mean and standard deviation of the healthy population data, k is the risk sensitivity coefficient, which can be configured and adjusted, and α is the basic risk influencing factor, which controls the threshold adjustment range.

[0025] The body fluctuation analysis unit 30 is used to collect family image data, perform behavior analysis on the elderly based on the family image data, determine the elderly's physical condition based on the behavior analysis results, and perform normal fluctuation trend analysis based on the physical condition combined with historical physical data; The body fluctuation analysis unit 30 includes an image acquisition module and a trend analysis module; The image acquisition module is used to collect indoor image data through home smart devices and aggregate the collected indoor image data into home image data; Image data acquisition: The camera equipment deployed in the room collects indoor image frames at preset time intervals; Family image data aggregation: Integrate continuously acquired image frames into a time-series image sequence to form complete family image data, classify the image sequence, distinguish different living scenes, and annotate the behavior of the elderly in the image.

[0026] The trend analysis module is used to analyze the behavior of the elderly based on family image data, obtain the behavior type of the elderly, and then analyze the physical state based on the behavior type and basic physical information to obtain the elderly's current physical state. The physical state is then combined with historical physical data to perform normal fluctuation trend analysis to obtain the normal fluctuation trend corresponding to each type of physical data. The specific steps are as follows: Behavior type analysis: Use computer vision algorithms to identify the elderly person's movements and positions in the image, match the identification results with predefined behavior templates (such as walking, sitting, lying down, falling, etc.), and determine the elderly person's current behavior type based on the matching results; Physical status analysis: Correlate behavior types with basic physical information of the elderly (such as age, medical history, exercise ability, etc.), and infer the physical status of the elderly (such as health, fatigue, discomfort, etc.) through behavioral characteristics; Normal fluctuation trend analysis: Combine historical physical data (such as heart rate, blood pressure, activity level, etc.) to build a time series model. Based on the current physical state and historical data patterns, the fluctuation range and trend of each physical data type under normal circumstances are predicted.

[0027] The body fluctuation analysis unit 30 obtains a collection time stamp of the body data and divides the body data into real-time body data and historical body data according to the collection time stamp; The collection timestamps adjacent to the real time are regarded as real-time body data, and the others are regarded as historical body data.

[0028] The body abnormality marking unit 40 is used to perform abnormal comparison on each type of real-time body data in combination with the risk status and the normal fluctuation trend, and mark the real-time body data of that type as abnormal according to the comparison result; The body abnormality marking unit 40 includes an abnormality comparison module and a marking summary module; The abnormality comparison module is used to perform real-time fluctuation analysis on each type of real-time body data, and compare the real-time fluctuation with the normal fluctuation trend. When the real-time fluctuation exceeds the normal fluctuation trend, it is marked as abnormal. Conversely, if the real-time fluctuation does not exceed the normal fluctuation trend, it will continue to be monitored. The specific steps are as follows: Real-time fluctuation analysis: For each type of real-time physical data (such as heart rate and blood pressure), the deviation between the current value and the historical average value (such as minute-level and hour-level fluctuations) is calculated, and real-time fluctuation characteristics such as fluctuation amplitude, frequency, and trend direction (upward / downward) are extracted. Normal fluctuation trend comparison: Call the normal fluctuation trend of this type of data in the trend analysis module, compare the real-time fluctuation characteristics with the normal fluctuation trend, and determine whether it exceeds the normal range; Abnormal flag trigger: If the real-time fluctuation exceeds the normal fluctuation trend range, it is marked as abnormal. Otherwise, continue to monitor subsequent real-time data. The formula is as follows: ; Where Δx(t) is the real-time fluctuation (i.e., change) at time point t, x(t) is the real-time body data at time point t, x(t-1) is the real-time body data at time point t-1, and NR(t) is the normal range of the data at time point t; If Δx(t)>NR(t), it is marked as abnormal; If Δx(t)≤NR(t), continue monitoring.

[0029] Based on real-time fluctuations and real-time body data, predictive analysis of body data is performed. The predicted body data is compared with the risk status. When the predicted body data is in a risky state, it is marked as abnormal. Conversely, when the predicted body data is not in a risky state, it is continued to be monitored. The specific steps are as follows: Physical data prediction analysis: Based on real-time fluctuations and real-time physical data, use time series prediction models (such as ARIMA and LSTM) to predict data values ​​for a period of time in the future (such as heart rate in the next 30 minutes); Risk status comparison (predicted value verification): Compare the predicted data value with the preset risk status threshold. If the predicted value touches the risk status, it is marked as abnormal; otherwise, continue monitoring.

[0030] The tagging and summarizing module is used to perform abnormal summary statistics on the marked real-time body data.

[0031] Taking into account both fluctuation exceeding the limit and predicted risk, abnormal data are combined and marked, and the abnormal type and time are recorded.

[0032] The rescue call management unit 50 includes an allocation setting module, a weight comparison module, and a collection reminder module; The allocation setting module is used to assign brain disease weights to each type of physical data combined with basic physical information, obtain the weight of each type of physical data on the onset of brain disease, and set abnormal thresholds based on the corresponding brain disease and basic physical information; Combining the elderly's basic physical information (such as age and medical history) and brain disease types, we assign a disease-related weight to each physical data type (such as blood pressure and heart rate) (the higher the weight, the greater the impact of the data on the disease). Based on the severity of the brain disease and basic physical information, we set an abnormality threshold (for example, emergency assistance is triggered when the total weight reaches 60%). The formula is as follows: ; Among them, J is the disease type, G is the basic information, S is the data type, and f is the weight distribution function based on medical knowledge or machine learning.

[0033] The weight comparison module is used to compare the abnormal real-time body data of the abnormal summary statistics with the corresponding weight and the abnormal threshold. When the weight occupied by the abnormal real-time body data statistics is greater than the abnormal threshold, the community medical center is called for rescue and examination. Conversely, when the weight occupied by the abnormal real-time body data statistics is less than the abnormal threshold, the collection reminder module is entered; ; Among them, W total is the sum of the weights of abnormal body data, and i∈Abnormality is the body data marked as abnormal.

[0034] Count the total weight of all current abnormal real-time body data and compare the total weight of abnormal data with the abnormal threshold; If the sum of the weights is greater than the abnormal threshold, a direct call will be made to the community medical center; If the total weight is less than or equal to the abnormal threshold, the collection reminder process will begin.

[0035] The collection reminder module is used to obtain the body data types that have not been collected in real time based on the real-time body data type. At the same time, based on the difference between the weight of the collected quantity and the abnormal threshold, the uncollected body types associated with the abnormal actual review body data are selected and sorted according to the proportion weight. The elderly are reminded to go to the home smart device corresponding to the highest-ranked uncollected body type to collect body data. If the elderly are unable to complete the collection, the community medical center is called for rescue and examination. After the elderly complete the body data collection, the weight is updated and compared with the abnormal threshold for a second time. The specific steps are as follows: Filtering of uncollected data types: Based on the collected real-time data types, determine the uncollected data types (e.g., blood pressure and heart rate have been collected, but blood oxygen and blood glucose have not been collected), and filter the uncollected types that are highly correlated with the current abnormal data (e.g., when blood pressure is abnormal, blood oxygen and blood glucose are prioritized); Sorting of uncollected data: Sort the uncollected data in descending order based on their weight ratio to generate a collection priority list; Collection reminder and response processing: Remind the elderly to use the corresponding smart devices to collect data in order of priority. If the elderly do not complete the collection or the timeout occurs, call the community medical center; if the collection is completed, update the weight sum and re-compare it with the threshold. The formula is as follows: ; Among them, Pr j To indicate the priority of the jth type of uncollected data, the larger the value, the higher the collection order. j is the association weight of the jth uncollected data type to brain disease, R j is the correlation between the jth uncollected data type and the current abnormal data, ∑(z j , R j ) is the sum of the combined importance of all uncollected data types and is used to normalize the numerator to the interval [0,1] to make the priorities comparable.

[0036] The number of collected data in the collection reminder module is the total number of body types corresponding to the real-time body data, and the number of uncollected data represents the total number of body types that can be collected by the home smart device excluding the body types corresponding to the number of collected data.

[0037] The above shows and describes the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The above embodiments and descriptions are merely preferred examples of the present invention and are not intended to limit the present invention. Various changes and improvements may be made to the present invention without departing from the spirit and scope of the present invention. Such changes and improvements fall within the scope of the present invention. The scope of protection claimed in the present invention is defined by the appended claims and their equivalents.

Claims

1. A remote home care detection system for the elderly in the community, characterized by: It includes a connection establishment unit (10), an elderly risk analysis unit (20), a body fluctuation analysis unit (30), a body abnormality marking unit (40), and a rescue call management unit (50); The connection establishment unit (10) is used to connect the home smart device and the community medical center, and to perform correlation analysis on the body data type collected by the home smart device in combination with brain diseases; The elderly risk analysis unit (20) is used to obtain physical information of the elderly, determine physical data related to brain diseases based on the physical information, and set a risk status for the physical data based on the physical information; The body fluctuation analysis unit (30) is used to collect family image data, perform behavior analysis on the elderly based on the family image data, determine the elderly's physical condition based on the behavior analysis results, and perform normal fluctuation trend analysis based on the physical condition combined with historical body data; The body abnormality marking unit (40) is used to perform abnormal comparison on each type of real-time body data in combination with the risk status and the normal fluctuation trend, and mark the real-time body data of that type as abnormal according to the comparison result; The rescue call management unit (50) is used to assign brain disease weights and abnormal thresholds to each type of body data in combination with body information, and then compare the weights of the abnormally marked real-time body data in combination with the abnormal thresholds, and select the body data type with the highest correlation when the real-time body data is abnormal and not collected to collect the data and remind the elderly.

2. A remote home care detection system for community elderly people according to claim 1, characterized in that: The connection establishment unit (10) establishes an intelligent management module through a home network terminal, and then establishes a connection between the intelligent management module and the home smart device, thereby obtaining the physical data collected by the home smart device on the elderly, and at the same time establishes a connection between the intelligent management module and the community medical center, thereby sending an emergency notification and the home location to the community medical center, so that the community medical center can provide assistance to the elderly.

3. The remote home care detection system for the elderly in the community according to claim 1 is characterized by: The connection establishment unit (10) includes a correlation analysis module; The correlation analysis module is used to identify the type of body data collected by home smart devices, obtain the data type corresponding to each body data, and then obtain the data type associated with each brain disease through medical atlas, so as to perform brain disease-related matching on each body data type of home smart devices.

4. The remote home care detection system for the elderly in the community according to claim 1 is characterized by: The elderly risk analysis unit (20) includes a body information acquisition module and a risk setting module; The body information acquisition module is used to collect body information related to the elderly from the home manager, identify brain diseases based on the relevant body information, and obtain the corresponding brain diseases and basic body information of the elderly based on the identification results; The risk setting module is used to set the risk status according to the brain diseases and basic physical information corresponding to the elderly combined with the physical data collected by home smart devices, and set the corresponding risk status for each type of physical data.

5. The remote home care detection system for community elderly people according to claim 1 is characterized by: The body fluctuation analysis unit (30) includes an image acquisition module and a trend analysis module; The image acquisition module is used to collect indoor image data through home smart devices and aggregate the collected indoor image data into home image data; The trend analysis module is used to perform behavioral analysis on the elderly based on family image data, obtain the behavior type the elderly are currently in, and then perform physical condition analysis based on the behavior type combined with basic physical information to obtain the elderly's current physical condition, and then perform normal fluctuation trend analysis on the physical condition combined with historical physical data to obtain the normal fluctuation trend corresponding to each type of physical data.

6. The remote home care detection system for the elderly in the community according to claim 1 is characterized by: The body fluctuation analysis unit (30) obtains a collection time stamp of the body data and divides the body data into real-time body data and historical body data according to the collection time stamp; The collection timestamps adjacent to the real time are regarded as real-time body data, and the others are regarded as historical body data.

7. The remote home care detection system for community elderly people according to claim 1 is characterized by: The body abnormality marking unit (40) includes an abnormality comparison module and a marking summary module; The abnormality comparison module is used to perform real-time fluctuation analysis on each type of real-time body data, compare the real-time fluctuation with the normal fluctuation trend, and mark it as abnormal when the real-time fluctuation exceeds the normal fluctuation trend. Otherwise, the real-time fluctuation does not exceed the normal fluctuation trend, and the monitoring is continued; Perform predictive analysis of body data based on real-time fluctuations combined with real-time body data, and compare the predicted body data with the risk status. If the predicted body data is in a risky state, it will be marked as abnormal. Conversely, if the predicted body data is not in a risky state, it will continue to be monitored. The marking summary module is used to perform abnormal summary statistics on the marked real-time body data.

8. The remote home care detection system for the elderly in the community according to claim 1 is characterized by: The rescue call management unit (50) includes an allocation setting module, a weight comparison module, and a collection reminder module; The allocation setting module is used to allocate brain disease weights to each type of physical data in combination with basic physical information, obtain the weight of each type of physical data for the onset of brain disease, and set abnormal thresholds according to the corresponding brain disease and basic physical information; The weight comparison module is used to compare the abnormal real-time body data of the abnormal summary statistics with the corresponding weight and the abnormal threshold. When the weight occupied by the abnormal real-time body data statistics is greater than the abnormal threshold, the community medical center is called for rescue and examination. On the contrary, when the weight occupied by the abnormal real-time body data statistics is less than the abnormal threshold, the collection reminder module is entered; The collection reminder module is used to obtain the body data type that has not been collected in real time based on the real-time body data type, and at the same time, according to the difference between the weight of the collected quantity and the abnormal threshold, select the uncollected body type associated with the abnormal actual review body data and sort it according to the proportion weight, and remind the elderly to go to the home smart device corresponding to the highest-ranked uncollected body type to collect body data. When the elderly are unable to complete the collection, call the community medical center for rescue and examination. After the elderly complete the body data collection, the weight is updated and compared with the abnormal threshold for a second time.

9. The remote home care detection system for community elderly people according to claim 8, characterized in that: The number of collected data in the collection reminder module is the total number of body types corresponding to the real-time body data, and the number of uncollected data represents the total number of body types that the home smart device can collect excluding the body types corresponding to the number of collected data.

Images