A community elderly care service management method based on big data

Abstract

The present invention relates to the technical field of big data analysis and community elderly care service management, and discloses a community elderly care service management method based on big data, including: collecting health data and behavior data of community elderly people in real time through Internet of Things devices and sensors; using big data analysis technology to perform data preprocessing and feature extraction on the health data and the behavior data to obtain key features; establishing a prediction model based on a decision tree algorithm, and training the prediction model; inputting the key features into the trained prediction model to predict the future health risks of community elderly people, and outputting the prediction results; the prediction results are displayed in the form of health reports, and personalized health management plans and services are recommended for community elderly people. The present invention solves the problems of insufficient data processing capabilities, low accuracy of prediction models, and lack of personalized services in the prior art, and improves the quality of community elderly care service management.

Description

Technical Field

[0001] The present invention relates to the technical field of big data analysis and community elderly care service management, and in particular to a community elderly care service management method based on big data. Background Art

[0002] With the aging of the population, the demand for community elderly care services is increasing. However, the existing community elderly care service management methods have many shortcomings, such as information islands, slow service response, lack of personalized services, etc. The current management system often cannot make full use of big data technology, resulting in the inability to provide efficient and accurate services; therefore, there is an urgent need for a community elderly care service management method based on big data to improve service quality and management efficiency.

[0003] The rapid development of modern information technology, especially the rise of big data and the Internet of Things (IoT), has brought new opportunities for community elderly care services. Through IoT devices, communities can monitor the health status and daily behaviors of the elderly in real time and collect a large amount of data; these data not only include the elderly’s physical health indicators, such as heart rate, blood pressure, body temperature, etc., but also cover their daily activities, such as walking distance, sleep quality, and eating habits; the accumulation of these data provides a basis for a deep understanding of the health status of the elderly. However, how to effectively process and analyze these huge amounts of data to achieve accurate predictions of the health status of the elderly and personalized services is still a challenge that needs to be solved.

[0004] Most of the existing community elderly care service management systems rely on traditional health monitoring methods, lack advanced data analysis capabilities, and find it difficult to fully utilize the massive amounts of data collected. In addition, these systems can usually only provide limited health advice and are unable to develop personalized health management plans based on the specific health conditions and living habits of the elderly. Summary of the invention

[0005] The purpose of this section is to summarize some aspects of the embodiments of the present invention and briefly introduce some preferred embodiments. Some simplifications or omissions may be made in this section and the abstract of the specification and the title of the invention of this application to avoid blurring the purpose of this section, the abstract of the specification and the title of the invention, and such simplifications or omissions cannot be used to limit the scope of the present invention.

[0006] In view of the above existing problems, the present invention is proposed.

[0007] Therefore, the technical problem solved by the present invention is that the existing community elderly care service management technology has the problems of insufficient data processing capability, low prediction model accuracy and lack of personalized services.

[0008] In order to solve the above technical problems, the present invention provides the following technical solutions: through IoT devices and sensors, health data and behavior data of the elderly in the community are collected in real time;

[0009] Using big data analysis technology to perform data preprocessing and feature extraction on the health data and the behavior data to obtain key features;

[0010] Establishing a prediction model based on a decision tree algorithm, and performing learning and training on the prediction model;

[0011] Inputting the key features into the trained prediction model to predict the future health risks of the elderly in the community, and outputting the prediction results;

[0012] The prediction results are presented in the form of health reports, recommending personalized health management plans and services for community elderly people.

[0013] As a preferred solution of the community elderly care service management method based on big data described in the present invention, the Internet of Things devices include but are not limited to heart rate monitors, blood pressure monitors, blood glucose meters, thermometers, respiratory rate monitors, activity trackers, cameras, and GPS positioning devices, which are used to detect the physiological indicators, movement trajectories, and behavior patterns of the elderly in the community;

[0014] The sensors include but are not limited to motion sensors, temperature and humidity sensors, and smoke detectors, which are used to detect the activities of the elderly in the community and environmental changes.

[0015] As a preferred solution of the community elderly care service management method based on big data described in the present invention, the health data at least includes heart rate, blood pressure, blood sugar, body temperature, weight, respiratory rate, sleep duration and depth;

[0016] The behavior data at least includes the number of steps, exercise time, GPS coordinates, temperature value, humidity value, smoke concentration value of the environment, frequency of daily activities, frequency of abnormal behavior, diet and medication intake.

[0017] As a preferred solution of the community elderly care service management method based on big data described in the present invention, data preprocessing is performed on the health data and the behavior data, including:

[0018] Cleaning the health data and the behavior data for missing values, abnormal values, and duplicate values;

[0019] Synchronize the cleaned health data and behavior data to ensure that the timestamps of the data are consistent;

[0020] Align the data in the same time period to form a unified time series;

[0021] The health data and the behavior data with consistent timestamps and unified time series are normalized to obtain numerical data.

[0022] As a preferred solution of the community elderly care service management method based on big data described in the present invention, the feature extraction can be expressed by the following mathematical formula:

[0023] Health data characteristics:

[0024]

[0025] Behavioral data characteristics:

[0026]

[0027] Environmental data characteristics:

[0028]

[0029] Environmental impact characteristics:

[0030]

[0031] Behavioral pattern characteristics:

[0032]

[0033] Among them, t0 and t n are the start and end time of data collection, HR(t) is the heart rate data at time t, , are the mean and standard deviation of heart rate data, BP(t) is the blood pressure data at time t, , are the mean and standard deviation of blood pressure data, BS(t) is the blood sugar data at time t, , are the mean and standard deviation of blood glucose data, is the weight coefficient of health data features, S i is the i-th step number data, is the step feature extraction function, A i is the i-th motion time data, is the motion feature extraction function, L i is the i-th GPS coordinate data, is the position feature extraction function, is the weight coefficient of the behavior data feature, Ej is the jth environmental data, is the mean of the environmental data, is the weight coefficient of the environmental data feature, Temp(t) is the temperature data at time t, , are the mean and standard deviation of the temperature data, Hum(t) is the humidity data at time t, , are the mean and standard deviation of humidity data, Smoke(t) is the smoke concentration data at time t, , are the mean and standard deviation of the smoke concentration data, is the weight coefficient of the environmental data feature, D k is the kth daily activity data, is the feature extraction function for daily activities, Ab k is the kth abnormal behavior data, is the abnormal behavior feature extraction function, is the weight coefficient of abnormal and daily behavior data features.

[0034] As a preferred solution of the community elderly care service management method based on big data described in the present invention, a prediction model is established based on a decision tree algorithm, including:

[0035]

[0036] Among them, K is the final predicted health index, that is, the predicted value, N(t) is the health data feature, B is the behavior data feature, E is the environmental data feature, E(t) is the environmental impact feature, and D is the behavior pattern feature.

[0037] As a preferred solution of the community elderly care service management method based on big data described in the present invention, K is a real number, where:

[0038] The larger the K value, the better the health status and behavioral characteristics of the elderly;

[0039] The smaller the K value, the worse the health status and behavioral characteristics of the elderly.

[0040] As a preferred solution of the community elderly care service management method based on big data described in the present invention, the learning training adjusts the model parameters through cross-validation to obtain the best model parameter combination.

[0041] As a preferred solution of the community elderly care service management method based on big data described in the present invention, the health report at least includes the changing trend of health indicators, health risk assessment, and health recommendations for the elderly.

[0042] Beneficial effects of the present invention:

[0043] 1. By deploying IoT devices and sensors in the community, real-time collection of health and behavior data of the elderly can be achieved, and the health status and daily activity information of the elderly can be continuously and accurately obtained, providing a comprehensive data foundation;

[0044] 2. Through big data analysis technology, the collected health data and behavior data are preprocessed and feature extracted, which can effectively filter out noise data and extract key features that have a significant impact on health status, ensuring high quality and high dimensionality of the data;

[0045] 3. The prediction model was established using the decision tree algorithm to achieve efficient prediction of future health risks of the elderly in the community, providing a scientific basis for health management;

[0046] 4. Help community caregivers and the elderly themselves to understand their health status in a timely manner, identify potential health risks, and take corresponding preventive measures. This not only improves the scientific nature and foresight of health management, but also provides more comprehensive health protection for the elderly. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for describing the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative labor. Among them:

[0048] Figure 1 This is a flow chart of the community elderly care service management method based on big data shown in the present invention. DETAILED DESCRIPTION

[0049] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the drawings. It is obvious that the described embodiments are only part of the embodiments of the present invention, but not all of the embodiments.

[0050] Based on the embodiments of the present invention, all other embodiments obtained by ordinary persons in the art without making any creative work should fall within the scope of protection of the present invention.

[0051] In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention may also be implemented in other ways different from those described herein, and those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0052] According to an embodiment of the present invention, Figure 1The flowchart shown is a community elderly care service management method based on big data, which specifically includes the following steps:

[0053] S1. Collect health and behavior data of the elderly in the community in real time through IoT devices and sensors;

[0054] S2. Use big data analysis technology to preprocess and extract features from health data and behavior data to obtain key features;

[0055] S3. Establish a prediction model based on the decision tree algorithm and conduct learning and training on the prediction model;

[0056] S4, input the key features into the trained prediction model to predict the future health risks of the elderly in the community, and output the prediction results;

[0057] S5. The prediction results are presented in the form of health reports to recommend personalized health management plans and services for the elderly in the community.

[0058] In an optional embodiment, the IoT devices include, but are not limited to, heart rate monitors, blood pressure monitors, blood glucose meters, thermometers, respiratory rate monitors, activity trackers, cameras, and GPS positioning devices for detecting the physiological indicators, movement trajectories, and behavioral patterns of the elderly in the community.

[0059] In an optional embodiment, the sensors include but are not limited to motion sensors, temperature and humidity sensors, and smoke detectors, which are used to detect the activities of elderly people in the community and environmental changes.

[0060] As an example, health data includes at least heart rate, blood pressure, blood sugar, body temperature, weight, breathing rate, sleep duration and depth.

[0061] As an example, the behavior data includes at least the number of steps, exercise time, GPS coordinates, temperature value, humidity value, smoke concentration value of the environment, frequency of daily activities, frequency of abnormal behaviors, diet and medication intake.

[0062] It should be noted that the frequency and accuracy of data collection need to be set according to the specific situation. For example, for key health indicators such as heart rate and blood pressure, a higher collection frequency can be set (such as once per minute) to ensure that changes in the health status of the elderly can be monitored in real time; for monitoring of activity trajectories and behavior patterns, the collection frequency can be appropriately lowered (such as once per hour) to reduce the amount of data and system load. The embodiments of the present invention do not make a sole limitation on this.

[0063] Preferably, the real-time collection method not only improves the timeliness and accuracy of the data, but also reduces the errors and delays in manual data collection, ensuring the reliability and effectiveness of subsequent data analysis. Through real-time monitoring, abnormal conditions of the elderly can be discovered in time, providing early warning information for community caregivers, thereby achieving early intervention and reducing health risks.

[0064] Furthermore, the health data and behavior data are preprocessed, including:

[0065] Clean up missing values, outliers, and duplicate values ​​in health and behavioral data;

[0066] Synchronize the cleaned health data and behavior data to ensure the consistency of the data timestamps;

[0067] Align the data in the same time period to form a unified time series;

[0068] The health data and behavioral data with consistent timestamps and unified time series are normalized to obtain numerical data.

[0069] As an example, feature extraction can be expressed by the following mathematical formula:

[0070] Health data characteristics:

[0071]

[0072] Behavioral data characteristics:

[0073]

[0074] Environmental data characteristics:

[0075]

[0076] Environmental impact characteristics:

[0077]

[0078] Behavioral pattern characteristics:

[0079]

[0080] Among them, t0 and t n are the start and end time of data collection, HR(t) is the heart rate data at time t, , are the mean and standard deviation of heart rate data, BP(t) is the blood pressure data at time t, , are the mean and standard deviation of blood pressure data, BS(t) is the blood sugar data at time t, , are the mean and standard deviation of blood glucose data, is the weight coefficient of health data features, S i is the i-th step number data, is the step feature extraction function, A i is the i-th motion time data, is the motion feature extraction function, L i is the i-th GPS coordinate data, is the position feature extraction function, is the weight coefficient of the behavior data feature, Ej is the jth environmental data, is the mean of the environmental data, is the weight coefficient of the environmental data feature, Temp(t) is the temperature data at time t, , are the mean and standard deviation of the temperature data, Hum(t) is the humidity data at time t, , are the mean and standard deviation of humidity data, Smoke(t) is the smoke concentration data at time t, , are the mean and standard deviation of the smoke concentration data, is the weight coefficient of the environmental data feature, D k is the kth daily activity data, is the feature extraction function for daily activities, Ab k is the kth abnormal behavior data, is the abnormal behavior feature extraction function, is the weight coefficient of abnormal and daily behavior data features.

[0081] Preferably, after data preprocessing and feature extraction, the accuracy and reliability of the subsequent prediction model are significantly improved, while the data redundancy and computational complexity are reduced, which not only improves the response speed of the system, but also enhances the ability to predict health risks, making the prediction results more accurate.

[0082] Furthermore, a prediction model is established based on the decision tree algorithm, which can be expressed by the following formula:

[0083]

[0084] Among them, K is the final predicted health index, that is, the predicted value, N(t) is the health data feature, B is the behavior data feature, E is the environmental data feature, E(t) is the environmental impact feature, and D is the behavior pattern feature.

[0085] It should be noted that K is a real number, where:

[0086] The larger the K value, the better the health status and behavioral characteristics of the elderly;

[0087] The smaller the K value, the worse the health status and behavioral characteristics of the elderly.

[0088] In an optional embodiment, the learning training adjusts the model parameters through cross-validation to obtain the best model parameter combination.

[0089] It is not difficult to understand that the decision tree algorithm has high interpretability and flexibility, and can process multi-dimensional and multi-variable data. Through continuous learning and training, the model can be gradually optimized to improve the accuracy and stability of prediction. Therefore, through this step, the prediction model can output personalized risk assessment results based on the health status and behavioral characteristics of different elderly people, thereby improving the pertinence and effectiveness of health management.

[0090] As an example, the health report at least includes the changing trends of health indicators of the elderly, health risk assessment, and health recommendations.

[0091] In an optional embodiment, health advice includes dietary advice, exercise plans, medication reminders, and regular physical examinations. For example, for elderly people at risk of hypertension, a low-salt diet, moderate exercise, and regular blood pressure monitoring measures can be recommended; for elderly people who are less active, a daily moderate exercise plan can be developed to improve their physical fitness.

[0092] Preferably, the community management system can match appropriate elderly care service resources based on the health data and behavioral data of the elderly. These service resources may include medical support, psychological counseling, emergency response services, and daily life assistance. For example, when the health status of the elderly is abnormal, the nearest medical institution or doctor can be recommended in time for diagnosis and treatment. When an elderly person has an accident, the emergency response service can be quickly initiated to ensure the safety of the elderly.

[0093] In addition, health management plans and service recommendations can be regularly adjusted and optimized based on the health status and behavior patterns of the elderly. For example, by analyzing the health data of the elderly over a period of time and finding that their health status has improved significantly, exercise plans and dietary recommendations can be appropriately adjusted; when the elderly's activity range and habits change, appropriate service resources can be re-matched and recommended.

[0094] As a preferred embodiment, verification is carried out through experiments, in which the experimental subjects are six elderly people in the community, aged between 65 and 80 years old, half of each gender, all of whom have chronic diseases to varying degrees, such as hypertension, diabetes, heart disease, etc. These elderly people usually live in the community and have regular daily activities, but their health conditions need to be continuously monitored and managed. They have signed informed consent forms and are willing to participate in the experiment.

[0095] Experimental location: The experiment was conducted in a community elderly care service center, which was equipped with necessary medical and monitoring equipment and had professional medical staff on site.

[0096] Experimental equipment: Heart rate monitor (brand: Omron, model: HBP-9020): used to measure the heart rate of the elderly, once in the morning and once in the evening every day, and the elderly must be in a sitting position during the measurement;

[0097] Blood pressure monitor (brand: Omron, model: HEM-7120): used to measure blood pressure, once in the morning and once in the evening every day. The elderly need to sit still and relax during the measurement to avoid measurement errors.

[0098] Blood glucose meter (brand: Roche, model: Accu-Chek Active): Measure the blood glucose level of the elderly three times a day (morning on an empty stomach, 2 hours after lunch, and 2 hours after dinner), ensuring that the elderly have clean hands before measurement;

[0099] Thermometer (Brand: Braun, Model: ThermoScan 7): Used to measure body temperature every morning. Be sure to disinfect the measuring probe during use to ensure accurate measurement.

[0100] Activity tracker (Brand: Fitbit, Model: Charge 4): Worn on the wrist of the elderly, it records the number of steps, exercise time and sleep duration throughout the day;

[0101] Camera (brand: Logitech, model: C920 HD Pro): Installed in the elderly activity area to record daily activities and collect behavioral data;

[0102] GPS positioning device (brand: Garmin, model: eTrex 32x): used to record the location and activity trajectory of the elderly, so as to facilitate the analysis of their activity range and frequency;

[0103] Experimental procedures: Heart rate and blood pressure were measured in the morning and evening, blood sugar was measured three times a day, and body temperature was measured in the morning;

[0104] All data is automatically uploaded to the central database through IoT devices;

[0105] Activity trackers record the number of steps, exercise time and sleep duration of the elderly, cameras record daily activities, GPS devices record location and trajectory, and all data is transmitted to the central database in real time;

[0106] Environmental sensors record temperature, humidity and smoke concentration, and the data is uploaded to the central database in real time;

[0107] Handle missing values, outliers and duplicate values ​​to ensure data integrity and accuracy;

[0108] Synchronize all data to ensure that the timestamps of the data are consistent;

[0109] Normalize the data for subsequent analysis and modeling;

[0110] Calculate the feature values ​​of health data, behavioral data, and environmental data to provide input for prediction models;

[0111] The extracted key features are input into the trained prediction model to predict the future health risks of the elderly;

[0112] The prediction results will be presented in the form of a health report, which includes the changing trends of health indicators, health risk assessment and health recommendations.

[0113] The experimental results are as follows:

[0114] Table 1. Parameter comparison table

[0115]

[0116] Referring to Table 1, it can be intuitively seen that the method of the present invention has significant advantages in health data collection and analysis, behavior data monitoring and management, environmental data evaluation and feedback, etc. Compared with traditional methods, the method of the present invention provides more accurate health risk assessment and personalized health management solutions through big data analysis and prediction models, which not only improves the quality and efficiency of community elderly care services, but also provides important technical support for health management in an aging society.

[0117] For example, Elderly person 2 has relatively high blood pressure and blood sugar levels. The predictive model identifies his potential health risks and provides detailed health advice in the health report. Traditional methods may not be able to detect these potential problems in a timely manner. The method of the present invention achieves accurate prediction and personalized management of the health status of the elderly through real-time monitoring and big data analysis.

[0118] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A community elderly care service management method based on big data, characterized in that: include: Collect health and behavior data of the elderly in the community in real time through IoT devices and sensors; The health data and the behavior data are preprocessed and feature extracted using big data analysis technology to obtain key features; the feature extraction can be expressed by the following mathematical formula: Health data characteristics: Behavioral data characteristics: Environmental data characteristics: Environmental impact characteristics: Behavioral pattern characteristics: Among them, t0 and t n are the start and end time of data collection, HR(t) is the heart rate data at time t, , are the mean and standard deviation of heart rate data, BP(t) is the blood pressure data at time t, , are the mean and standard deviation of blood pressure data, BS(t) is the blood sugar data at time t, , are the mean and standard deviation of blood glucose data, is the weight coefficient of health data features, S i is the i-th step number data, is the step feature extraction function, A i is the i-th motion time data, is the motion feature extraction function, L i is the i-th GPS coordinate data, is the position feature extraction function, is the weight coefficient of the behavior data feature, Ej is the jth environmental data, is the mean of the environmental data, is the weight coefficient of the environmental data feature, Temp(t) is the temperature data at time t, , are the mean and standard deviation of the temperature data, Hum(t) is the humidity data at time t, , are the mean and standard deviation of humidity data, Smoke(t) is the smoke concentration data at time t, , are the mean and standard deviation of the smoke concentration data, is the weight coefficient of the environmental data feature, D k is the kth daily activity data, is the feature extraction function for daily activities, Ab k is the kth abnormal behavior data, is the abnormal behavior feature extraction function, is the weight coefficient of abnormal and daily behavior data features; Build a prediction model based on the decision tree algorithm, including: Among them, K is the final predicted health index, that is, the predicted value, N (t) is the health data feature, B is the behavior data feature, E is the environmental data feature, E (t) is the environmental impact feature, and D is the behavior pattern feature; K is a real number, where: The larger the K value, the better the health status and behavioral characteristics of the elderly; The smaller the K value, the worse the health status and behavioral characteristics of the elderly; and performing learning and training on the prediction model; Inputting the key features into the trained prediction model to predict the future health risks of the elderly in the community, and outputting the prediction results; The prediction results are presented in the form of health reports, recommending personalized health management plans and services for community elderly people.

2. The community elderly care service management method based on big data according to claim 1 is characterized in that: The IoT devices include heart rate monitors, blood pressure monitors, blood glucose meters, thermometers, respiratory rate monitors, activity trackers, cameras, and GPS positioning devices to detect the physiological indicators, movement trajectories, and behavior patterns of the elderly in the community; The sensors include but are not limited to motion sensors, temperature and humidity sensors, and smoke detectors, which are used to detect the activities of the elderly in the community and environmental changes.

3. The community elderly care service management method based on big data according to claim 1 is characterized in that: The health data at least includes heart rate, blood pressure, blood sugar, body temperature, weight, respiratory rate, sleep duration and depth; The behavior data at least includes the number of steps, exercise time, GPS coordinates, temperature value, humidity value, smoke concentration value of the environment, frequency of daily activities, frequency of abnormal behavior, diet and medication intake.

4. The community elderly care service management method based on big data according to claim 3 is characterized in that: The health data and the behavior data are preprocessed, including: Cleaning the health data and the behavior data for missing values, abnormal values, and duplicate values; Synchronize the cleaned health data and behavior data to ensure that the timestamps of the data are consistent; Align the data in the same time period to form a unified time series; The health data and the behavior data with consistent timestamps and unified time series are normalized to obtain numerical data.

5. The community elderly care service management method based on big data according to claim 1 is characterized in that: The learning training adjusts the model parameters through cross-validation to obtain the best model parameter combination.

6. The community elderly care service management method based on big data according to claim 1 is characterized in that: The health report at least includes the changing trends of health indicators, health risk assessment, and health recommendations for the elderly.

Images