Method and system for fusing and recommending elderly nursing activity indexes

Abstract

The invention belongs to the technical field of elderly nursing, and relates to an elderly nursing activity index fusion recommendation method and system. Comprising a data acquisition module which is deployed in each key area of an elderly nursing place, comprises a plurality of sensors and a data input interface, and is used for acquiring multi-dimensional data such as physiological indexes, psychological states, activity ability and social communication of the elderly, adding timestamps to the acquired data, and sending the data to the elderly nursing place; preliminarily classifying the data and then transmitting the data to a data processing module through a data transmission network; the data transmission network is used for transmitting the data acquired by the data acquisition module; according to the method, comprehensive evaluation of the individual condition of the old people is realized by fusing multi-dimensional indexes such as physiology, psychology, activity ability and social communication, the accuracy and pertinence of old people nursing activity recommendation are improved, and the limitation of single-index recommendation is avoided. A dynamic adjustment mechanism is established, and a recommendation scheme can be optimized in time according to the real-time state and the nursing effect of the old people.

Description

Technical Field

[0001] The present invention belongs to the technical field of elderly care, and in particular, relates to a method and system for integrating and recommending elderly care activity indicators. Background Art

[0002] With the aging of the population, the demand for elderly care is growing. Currently, recommendations for elderly care activities are mostly based on a single indicator or simple rules, lacking a comprehensive consideration of multi-dimensional indicators such as the elderly's physiological, psychological, and social aspects. For example, recommending care activities based solely on the elderly's age or medical condition fails to accurately reflect individual differences, resulting in recommended care activities that are either too conservative and fail to meet the elderly's actual needs, or exceed the elderly's physical tolerance, posing a safety hazard. Furthermore, existing recommendation methods lack a dynamic adjustment mechanism and cannot timely optimize recommended plans based on the elderly's real-time status and care effectiveness, affecting the quality and efficiency of elderly care.

[0003] There is an urgent need for a method and system for recommending elderly care activities that can integrate multi-dimensional indicators and dynamically adjust recommendation plans. To this end, the present invention provides a method and system for recommending elderly care activity indicators through integration. Summary of the Invention

[0004] The purpose of the present invention is to provide a method and system for fusion recommendation of elderly care activity indicators, which solves the problems existing in the prior art.

[0005] The purpose of the present invention can be achieved through the following technical solutions:

[0006] A system for integrating and recommending elderly care activity indicators, comprising:

[0007] The data acquisition module is deployed in key areas of elderly care facilities and contains multiple sensors and data input interfaces. It is used to collect multi-dimensional data such as the elderly's physiological indicators, psychological state, mobility, social interaction, etc., add timestamps to the collected data, and transmit the data to the data processing module through the data transmission network after preliminary classification;

[0008] Data transmission network, used to transmit data collected by the data acquisition module;

[0009] The data processing module receives data transmitted by the data acquisition module, performs multi-dimensional indicator fusion processing, weight calculation, and similarity analysis on the data, establishes a nursing activity recommendation model, detects data anomalies and generates adjustment instructions, and adjusts the recommendation strategy and weight parameters according to the recommendation effect and data changes;

[0010] The recommendation decision module makes nursing activity recommendation decisions and generates recommendation plans based on the processing results of the data processing module and the personalized needs and preferences of the elderly;

[0011] The user interaction module is used for the elderly and their caregivers to interact with the system, input personalized needs, view recommended solutions, and provide feedback on usage effects.

[0012] Preferably, the physiological indicator sensor includes one or more of a heart rate sensor, a blood pressure sensor, and a blood oxygen saturation sensor, and the activity monitoring device includes one or more of a gait monitoring sensor and a balance ability detection device.

[0013] Preferably, the multi-dimensional indicator fusion processing adopts a weighted fusion algorithm, and the initial weight value is determined based on expert knowledge and historical data in the field of elderly care. The weight of physiological indicators is 0.3-0.5, the weight of psychological indicators is 0.2-0.4, the weight of activity ability indicators is 0.1-0.3, and the weight of social interaction indicators is 0.05-0.2.

[0014] Preferably, the similarity analysis uses a cosine similarity algorithm, and the formula is: Where A is the fusion data vector of the elderly, and B is the characteristic indicator vector of nursing activities.

[0015] A method for integrating and recommending elderly care activity indicators includes the following steps:

[0016] Data collection and initial classification: Deploy a variety of sensors and data input interfaces in elderly care facilities to collect multi-dimensional data such as the elderly's physiological indicators, psychological state, mobility, social interaction, etc., perform initial classification based on data characteristics and elderly care needs, and transmit the data to the data processing module through the data transmission network;

[0017] Data processing module processing flow: After receiving the data, the data processing module integrates the multi-dimensional indicators, calculates the similarity with the characteristic indicators of each activity in the elderly care activity database, establishes a nursing activity recommendation model, determines the recommendation candidate set based on the similarity results and the preset threshold, and regularly adjusts the parameters and weights of the recommendation model;

[0018] Abnormal data processing mechanism: By setting abnormal thresholds and time intervals for inspection, abnormal data is detected and handled according to different situations using early warning, interpolation completion, inference completion and other methods;

[0019] Recommendation decision and adjustment: The recommendation decision module generates a recommendation plan based on the recommendation candidate set and the personalized needs of the elderly, evaluates the recommendation effect through the recommendation effect evaluation system, and adjusts the recommendation strategy and weight parameters according to the evaluation results.

[0020] Preferably, the abnormal threshold is determined based on the standard deviation and interquartile range of historical data, and the formula is T=k·σ, where k is the coefficient, σ is the standard deviation of historical data, and the data time interval Δt exceeds the maximum allowable interval of the corresponding acquisition frequency. It is judged as abnormal when , and f is the acquisition frequency.

[0021] Preferably, the interpolation complement adopts linear interpolation or Lagrange interpolation algorithm, and the linear interpolation formula is: Where t1, t2, and t3 are time points, and d1, d2, and d3 are data values ​​at the corresponding time points.

[0022] Preferably, the recommendation effect evaluation system evaluates the recommendation plan from the aspects of improvement of physiological indicators, improvement of psychological state, satisfaction of the elderly, etc. When the evaluation results do not meet expectations, the recommendation strategy and weight parameters are adjusted.

[0023] Beneficial effects of the present invention: The present invention achieves a comprehensive assessment of the individual conditions of the elderly by integrating multi-dimensional indicators such as physiology, psychology, activity ability, and social interaction, improves the accuracy and pertinence of recommendations for elderly care activities, and avoids the limitations of single indicator recommendations. A dynamic adjustment mechanism has been established, which can timely optimize the recommendation plan according to the real-time status and care effect of the elderly, ensuring that the recommended care activities not only meet the actual needs of the elderly, but also do not exceed their physical tolerance, thereby improving the quality and safety of elderly care. The perfect abnormal data detection and processing mechanism can timely discover and process data anomalies, ensure the integrity and accuracy of the data, and provide strong support for scientific and reasonable recommendation decisions. The personalized recommendation decision and user interaction module fully consider the personalized needs and preferences of the elderly, improve the elderly's participation and satisfaction in care activities, and help improve the overall effect of elderly care. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] 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 work.

[0025] Figure 1 This is a system block diagram of the present invention for integrating and recommending elderly care activity indicators;

[0026] Figure 2 This is a flow chart of a method for integrating and recommending elderly care activity indicators according to the present invention. DETAILED DESCRIPTION

[0027] 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 any creative efforts shall fall within the scope of protection of the present invention.

[0028] See also Figure 1 As shown, the present invention is a system for integrating and recommending elderly care activity indicators.

[0029] Data Collection Module: Deployed in key areas of elderly care facilities, this module contains multiple sensors and data input interfaces. It collects multi-dimensional data on the elderly's physiological indicators, psychological state, mobility, social interactions, and other aspects. It also adds timestamps to the collected data, performs preliminary classification, and transmits it to the data processing module via a data transmission network. Physiological indicator sensors include heart rate sensors, blood pressure sensors, and blood oxygen saturation sensors; mobility monitoring devices include gait monitoring sensors and balance monitoring devices.

[0030] Data transmission network: used to transmit data collected by the data acquisition module, including 5G, Wi-Fi or dedicated networks.

[0031] Data processing module: As the core processing unit, it receives data transmitted by the data acquisition module, performs multi-dimensional indicator fusion processing, weight calculation, and similarity analysis on the data, establishes a nursing activity recommendation model, detects data anomalies and generates adjustment instructions, and adjusts the recommendation strategy and weight parameters according to the recommendation effect and data changes.

[0032] Recommendation decision module: Based on the processing results of the data processing module and combined with the personalized needs and preferences of the elderly, it makes nursing activity recommendation decisions and generates a detailed recommendation plan.

[0033] User interaction module: used for the elderly and their caregivers to interact with the system, including inputting personalized needs, viewing recommended solutions, and providing feedback on usage effects.

[0034] See also Figure 2 As shown, a method for integrating and recommending elderly care activity indicators

[0035] Data Collection and Initial Classification: Various sensors and data input interfaces are deployed in key areas of elderly care facilities to collect and initially classify multi-dimensional data based on data characteristics and elderly care needs. Physiological indicators, such as heart rate and blood pressure, are collected in real time or at a scheduled frequency. Psychological data is collected daily or weekly through questionnaires or interactive tests. Activity data is collected per activity or at a scheduled frequency using devices such as gait monitoring. Social interaction data is collected weekly or monthly through social activity records. The data collection module performs preliminary classification of the collected data and transmits it to the data processing module via the data transmission network.

[0036] Data processing module processing flow

[0037] Multi-dimensional indicator fusion: After receiving data, the data processing module integrates multi-dimensional indicators. Using a weighted fusion algorithm, it comprehensively calculates physiological, psychological, mobility, and social interaction indicators. Initial weights are determined based on expert knowledge and historical data in the field of elderly care. For example, physiological indicators are weighted at 0.4, psychological indicators at 0.3, mobility indicators at 0.2, and social interaction indicators at 0.1.

[0038] Similarity analysis: A database of elderly care activities is established, which contains the characteristic indicators of various care activities and the scope of applicable populations. For each elderly person's fused data, the similarity between it and the characteristic indicators of each care activity in the database is calculated. The similarity calculation uses the cosine similarity algorithm, and the formula is: Where A is the fusion data vector of the elderly, and B is the characteristic indicator vector of nursing activities.

[0039] Recommendation model establishment and adjustment: Based on the similarity analysis results, a nursing activity recommendation model is established. When the similarity exceeds a preset threshold (e.g., 0.7), the corresponding nursing activity is included in the recommendation candidate set. Simultaneously, the parameters and weights of the recommendation model are regularly adjusted based on the elderly's nursing feedback and data changes. For example, if an elderly person's physiological indicators and psychological state improve significantly after participating in a recommended activity, the weight of the characteristic indicator corresponding to that activity is increased.

[0040] Abnormal data processing mechanism

[0041] Anomaly detection: Set an initial anomaly threshold for each data type. This threshold is determined based on statistical analysis of historical data, such as the standard deviation σ and interquartile range of historical data, which can be expressed as T = k·σ (k is a coefficient, adjusted according to data characteristics). If the deviation of a certain data from the historical mean exceeds the threshold, or the data time interval Δt exceeds the maximum allowable interval of the corresponding acquisition frequency, (f is the acquisition frequency), it is determined to be abnormal data.

[0042] Abnormal processing and data correction: For high-priority abnormal data, such as a sudden abnormal increase in heart rate, the data processing module immediately sends an early warning message to the nursing staff and suspends the recommendation of related high-risk nursing activities. For abnormal data with medium and low priority and a short missing time, an interpolation algorithm is used to complete it, such as linear interpolation, the formula is Where t1, t2, and t3 are time points, and d1, d2, and d3 are the data values ​​at the corresponding time points. For long-term abnormal data caused by reasons such as equipment failure, inference and completion are performed by combining historical data of the elderly and average data of similar populations.

[0043] Recommended decisions and adjustments

[0044] Recommendation plan generation: Based on the results of the data processing module, the recommendation decision module selects the most similar care activities from the recommended candidate set and generates a detailed recommendation plan, including activity content, intensity, and time schedule. At the same time, the individual needs and preferences of the elderly are taken into consideration. For example, if the elderly prefer outdoor activities, activities such as outdoor walking and gardening will be prioritized.

[0045] Evaluation and Adjustment of Recommendation Effectiveness: Establish a recommendation effectiveness evaluation system to assess recommended solutions based on improvements in physiological indicators, psychological well-being, and elderly satisfaction. If the evaluation results fall short of expectations, analyze the reasons and adjust the recommendation strategy and weighting parameters. For example, if a particular activity's recommendation performance is poor, reduce the weight of its corresponding characteristic indicator or remove the activity from the recommended candidate set.

[0046] The specific embodiments are as follows:

[0047] A system for integrating and recommending elderly care activity indicators includes a data acquisition module, a data transmission network, a data processing module, a recommendation decision module and a user interaction module, wherein the data acquisition module is connected to the data transmission network, the data transmission network is connected to the data processing module, the data processing module is connected to the recommendation decision module, and the recommendation decision module is connected to the user interaction module.

[0048] Data collection modules are deployed in key areas of elderly care centers, such as bedrooms, activity areas, and dining rooms. These modules are equipped with heart rate sensors, blood pressure sensors, and gait monitoring sensors. For example, in one elderly care center, heart rate and blood pressure sensors are installed in the bedrooms of residents to collect real-time physiological data. Gait monitoring carpets are laid in activity areas to collect data on the residents' mobility. Residents are asked to complete daily psychological questionnaires using devices such as tablets to collect psychological data. Caregivers also regularly record the residents' social interactions to collect data on their interactions.

[0049] After collecting data, the data acquisition module performs preliminary classification according to preset classification rules and transmits the data to the data processing module via the 5G network. After receiving the data, the data processing module integrates the multi-dimensional indicators and calculates the weighted sum of physiological indicators (weight 0.4), psychological indicators (weight 0.3), activity indicators (weight 0.2), and social interaction indicators (weight 0.1). The similarity between this integrated data and the characteristic indicators of each activity in the elderly care activity database is then calculated using the cosine similarity algorithm. Activities with a similarity greater than 0.7 are included in the recommendation candidate set.

[0050] In terms of abnormal data detection, if an elderly person's heart rate data deviates from the historical mean by more than two standard deviations, it is identified as abnormal data. The data processing module immediately sends an alert to the caregiver and suspends recommendations for high-intensity activities. For missing daily mental state data, if the missing time is short, linear interpolation is used to fill in the gaps.

[0051] The recommendation decision module selects the three most similar care activities from the candidate set and generates a recommended plan. For example, for an elderly person in good health and who enjoys socializing, it recommends outdoor walking, crafting, and group chatting, along with specific time and intensity recommendations. The elderly person and their caregiver review the recommended plans through the user interaction module and provide feedback on their performance. If the elderly person's heart rate and blood pressure remain within normal ranges and their psychological state improves after participating in the outdoor walking activity, the weight of the corresponding characteristic indicator for the outdoor activity is increased. If the elderly person has low interest in crafting and low satisfaction, the weight of this activity is reduced.

[0052] Throughout the specification, references to terms such as "one embodiment," "example," or "specific example" indicate that the specific features, structures, materials, or characteristics described in conjunction with that embodiment or example are included in at least one embodiment or example of the present invention. In this specification, schematic representations of these terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples.

[0053] The above content is merely an example and explanation of the concept of the present invention. Those skilled in the art may make various modifications or additions to the described specific embodiments or replace them in a similar manner. As long as they do not deviate from the concept of the invention or exceed the scope defined by the claims, they should all fall within the scope of protection of the present invention.

Claims

1. A system for integrating and recommending elderly care activity indicators, characterized by: include: The data acquisition module is deployed in key areas of elderly care facilities and contains multiple sensors and data input interfaces. It is used to collect multi-dimensional data such as the elderly's physiological indicators, psychological state, mobility, social interaction, etc., add timestamps to the collected data, and transmit the data to the data processing module through the data transmission network after preliminary classification; Data transmission network, used to transmit data collected by the data acquisition module; The data processing module receives data transmitted by the data acquisition module, performs multi-dimensional indicator fusion processing, weight calculation, and similarity analysis on the data, establishes a nursing activity recommendation model, detects data anomalies and generates adjustment instructions, and adjusts the recommendation strategy and weight parameters according to the recommendation effect and data changes; The recommendation decision module makes nursing activity recommendation decisions and generates recommendation plans based on the processing results of the data processing module and the personalized needs and preferences of the elderly; The user interaction module is used for the elderly and their caregivers to interact with the system, input personalized needs, view recommended solutions, and provide feedback on usage effects.

2. The system for fusion recommendation of elderly care activity indicators according to claim 1 is characterized by: The physiological indicator sensor includes one or more of a heart rate sensor, a blood pressure sensor, and a blood oxygen saturation sensor, and the activity monitoring device includes one or more of a gait monitoring sensor and a balance ability detection device.

3. The system for fusion recommendation of elderly care activity indicators according to claim 1 is characterized by: The multi-dimensional indicator fusion processing adopts a weighted fusion algorithm, and the initial weight value is determined based on expert knowledge and historical data in the field of elderly care. The weight of physiological indicators is 0.3-0.5, the weight of psychological indicators is 0.2-0.4, the weight of activity ability indicators is 0.1-0.3, and the weight of social interaction indicators is 0.05-0.

2.

4. The system for fusion recommendation of elderly care activity indicators according to claim 1 is characterized by: The similarity analysis adopts the cosine similarity algorithm, and the formula is: Where A is the fusion data vector of the elderly, and B is the characteristic indicator vector of nursing activities.

5. A method for fusion recommendation of elderly care activity indicators, applied to the system for fusion recommendation of elderly care activity indicators according to claims 1-4, characterized in that: The following steps are involved: Data collection and initial classification: Deploy a variety of sensors and data input interfaces in elderly care facilities to collect multi-dimensional data such as the elderly's physiological indicators, psychological state, mobility, social interaction, etc., perform initial classification based on data characteristics and elderly care needs, and transmit the data to the data processing module through the data transmission network; Data processing module processing flow: After receiving the data, the data processing module integrates the multi-dimensional indicators, calculates the similarity with the characteristic indicators of each activity in the elderly care activity database, establishes a nursing activity recommendation model, determines the recommendation candidate set based on the similarity results and the preset threshold, and regularly adjusts the parameters and weights of the recommendation model; Abnormal data processing mechanism: By setting abnormal thresholds and time intervals for inspection, abnormal data is detected and handled according to different situations using early warning, interpolation completion, inference completion and other methods; Recommendation decision and adjustment: The recommendation decision module generates a recommendation plan based on the recommendation candidate set and the personalized needs of the elderly, evaluates the recommendation effect through the recommendation effect evaluation system, and adjusts the recommendation strategy and weight parameters according to the evaluation results.

6. The method for fusion recommendation of elderly care activity indicators according to claim 5, characterized in that: The abnormal threshold is determined based on the standard deviation and interquartile range of historical data. The formula is T = k·σ, where k is the coefficient, σ is the standard deviation of historical data, and the data time interval Δt exceeds the maximum allowable interval of the corresponding acquisition frequency. It is judged as abnormal when , and f is the acquisition frequency.

7. The method for fusion recommendation of elderly care activity indicators according to claim 5, characterized in that: The interpolation completion adopts linear interpolation or Lagrange interpolation algorithm, and the linear interpolation formula is: Where t1, t2, and t3 are time points, and d1, d2, and d3 are data values ​​at the corresponding time points.

8. The method for fusion recommendation of elderly care activity indicators according to claim 5 is characterized by: The recommendation effect evaluation system evaluates the recommendation plan from the aspects of improvement of physiological indicators, improvement of psychological state, and satisfaction of the elderly. When the evaluation results do not meet expectations, the recommendation strategy and weight parameters are adjusted.

Images