Nursing home priority service response method, system and device based on digital twin

Abstract

The present invention relates to a nursing home priority service response method, system and device based on digital twins, comprising the following steps: collecting health data of the elderly and status data of caregivers, and preprocessing the data; constructing a dynamic model of the elderly's health status, a dynamic model of care resources and an environmental path model; coupling the above three models to form a digital twin model; calculating the service response priority based on the elderly's health score, waiting time decay and medical history characteristics; scheduling matching caregivers to respond to the service according to the service response priority; collecting the service execution results after the caregivers complete the task, and based on the service execution results, updating the digital twin model and dynamically adjusting the scheduling strategy. The present invention realizes real-time mapping and dynamic optimization of the health status of the elderly, the location of caregivers and the environmental path information in the nursing home, significantly improving the service response efficiency, resource scheduling accuracy and care quality.

Description

Technical Field

[0001] The present invention relates to the fields of smart elderly care and resource optimization scheduling, and in particular to a digital twin-based nursing home priority service response method, device and medium. Background Art

[0002] With the accelerating aging of the global population, the health management and elderly care service needs of the elderly are becoming increasingly complex, placing higher demands on the service quality and responsiveness of nursing homes. However, the traditional nursing home service model, which relies primarily on manual scheduling, struggles to meet real-time and precision requirements and has numerous limitations. Technological innovation is urgently needed to achieve intelligent upgrades.

[0003] Current health management systems often rely on single sensors (such as heart rate monitors and blood pressure monitors), which can only capture isolated physiological indicators and fail to fully reflect the dynamic changes in an elderly person's health. For example, heart rate monitoring alone cannot determine whether an elderly person is in an emergency situation, such as a fall. Furthermore, existing systems lack integrated analysis of the surrounding environment, making it difficult to comprehensively assess the impact of the elderly person's surroundings on their health. For example, a humid environment may increase the risk of falls for an elderly person, but a single sensor system cannot capture this correlation.

[0004] Traditional nursing home service response mechanisms mostly rely on fixed priority rules, ignoring the dynamic changes in caregivers' real-time location, skill matching, and the risk level of residents. This model leads to uneven resource allocation and delayed responses. For example, if a high-risk resident experiences a sudden health issue and there are no available caregivers nearby, the system cannot dynamically adjust task allocation based on actual conditions, resulting in extended response times. Furthermore, under this traditional model, caregiver workloads are difficult to monitor in real time, making it easy for some caregivers to become overloaded while others remain idle.

[0005] While the smart home sector has experimented with integrating various sensor technologies (such as Wi-Fi positioning and BLE sensors), their applications are often limited to health monitoring and fail to effectively integrate them into service scheduling decisions. For example, while a fall detection system can trigger a local alarm, it cannot be linked to a caregiver dispatch system, preventing emergency situations from being addressed promptly.

[0006] In recent years, multi-sensor integration technology has provided new solutions to the above problems by integrating multiple sensors such as heart rate, blood pressure, and activity monitoring sensors to comprehensively monitor the health of the elderly. However, existing technologies still have the following shortcomings:

[0007] Insufficient data fusion and processing capabilities make it impossible to accurately reflect the health status of the elderly in real time;

[0008] There is a lack of real-time monitoring and scheduling optimization mechanisms for nursing staff's dynamic information;

[0009] Data security and privacy protection have not been fully considered, making it difficult to meet the needs of multi-institutional collaboration.

[0010] Therefore, developing an intelligent service response mechanism that can integrate multimodal data, dynamically adjust service priorities, optimize resource scheduling, and ensure data security has become an urgent problem to be solved in the field of smart elderly care. Summary of the Invention

[0011] To this end, the technical problem to be solved by the present invention is to overcome the shortcomings of the existing technology. The present invention provides a nursing home priority service response method, device and medium based on digital twins, which realizes real-time mapping and dynamic optimization of the health status of the elderly in the nursing home, the location of caregivers and environmental path information, significantly improves the service response efficiency, resource scheduling accuracy and nursing quality, and provides a reliable and efficient solution for smart elderly care.

[0012] In the first aspect, to solve the above technical problems, the present invention provides a nursing home priority service response method based on digital twins, comprising the following steps:

[0013] S1: Collect health data of the elderly and status data of caregivers, and pre-process the data;

[0014] S2: Based on the preprocessed data, a dynamic model of the elderly’s health status and a dynamic model of nursing resources are constructed respectively. At the same time, an environmental path model is constructed based on the building layout data of the nursing home.

[0015] S3: Coupling the data of the elderly health status dynamic model, the nursing resource dynamic model, and the environmental path model to form a digital twin model;

[0016] S4: Calculate service response priority based on the health score, waiting time decay, and medical history characteristics of the elderly in the digital twin model;

[0017] S5: dispatching matching nursing staff to perform service response according to the service response priority;

[0018] S6: Collect the service execution results after the caregiver completes the task, and based on the service execution results, update the health score, caregiver status and service response priority in the digital twin model, and dynamically adjust the scheduling strategy.

[0019] In one embodiment of the present invention, in step S2, the dynamic model of the health status of the elderly is used to fuse multi-source data into a single health score, which is expressed as follows:

[0020] ;

[0021] represents the health score of the elderly at time t, To score abnormal heart rate, Score abnormal blood pressure. Score the fall sign, is the risk model;

[0022] Among them, the risk model The expression is as follows:

[0023] ;

[0024] A is the age coefficient, C is the weight of chronic diseases, E is the recent abnormal events, and D is the weight of new diseases.

[0025] In one embodiment of the present invention, in step S2, the dynamic nursing resource model is used to update the location, skill tags, and task queues of nursing staff in real time, and its expression is as follows:

[0026] ;

[0027] in, The location information of the caregiver, Skill labels for nursing staff; It is a task queue used to limit the maximum number of concurrent tasks; It is the working status parameter.

[0028] In one embodiment of the present invention, the environmental path model in step S2 is used to simulate the physical environment inside the nursing home to form a pre-calculated path time matrix; the pre-calculated path time matrix T is an N×N symmetric matrix, where N is the total number of key location nodes in the nursing home, and the matrix elements It represents the estimated travel time from location node A to B, and its calculation formula is:

[0029] ;

[0030] in, is the physical distance from location A to location B; v is the walking speed of the caregiver; The estimated travel time from point A to point B; Allow for additional time when switching between elevators or stairs; It is a dynamic correction item based on real-time data.

[0031] In one embodiment of the present invention, in step S4, the priority of the service demand is evaluated by integrating the health score, waiting time decay and medical history characteristics through a multi-factor priority algorithm; the calculation formula of the service response priority is as follows:

[0032] ;

[0033] in, For service response priority, is the health score, M is the medical history characteristics; is the waiting time decay function, .

[0034] In one embodiment of the present invention, in step S5, the step of dispatching a nursing staff to perform a service response includes:

[0035] S51: Based on the pre-calculated path time matrix, calculate the shortest path from the current location of different caregivers to the service location for each service demand;

[0036] S52: Based on the shortest path, nursing staff skill matching and task load, the most suitable nursing staff is dispatched for service response through a multi-objective optimization model;

[0037] S53: Identify tasks in the service response that are geographically close or can be completed consecutively in time;

[0038] S54: Evaluate the feasibility of merging tasks into a task package;

[0039] S55: Integrate the mergeable tasks into a task package, and recalculate the estimated completion time and path of the merged tasks;

[0040] S56: Update the task allocation of the nursing staff according to the merged task package.

[0041] In one embodiment of the present invention, the expression of the multi-objective optimization model is as follows:

[0042] ;

[0043] in, is the path time; For skill matching; For task load.

[0044] In one embodiment of the present invention, in step S6, the service execution result includes task completion status, actual response time, task execution quality, nursing staff feedback, system records and logs, and health status updates.

[0045] In the second aspect, in order to solve the above technical problems, the present invention provides a nursing home priority service response system based on digital twins, including:

[0046] The perception layer includes a physiological detection module and an environmental positioning module; the physiological detection module includes a heart rate sensor, a blood pressure sensor, and an activity monitoring sensor for real-time collection of the elderly's physiological indicators; the environmental positioning module is used to monitor and record the location and trajectory of the elderly and caregivers in the nursing home;

[0047] The edge computing layer is used to preprocess data and build a digital twin model to map the elderly's health status and nursing resource status in real time; the digital twin model includes a dynamic model of the elderly's health status, a dynamic model of nursing resources, and an environmental path model;

[0048] A decision-making layer, configured to dynamically calculate service response priorities and optimize resource allocation based on nursing staff's skills, location, and task load; the decision-making layer calculates service response priorities using a multi-factor priority algorithm;

[0049] The execution layer is used to send service instructions to nursing staff;

[0050] The feedback layer is used to collect feedback data, iterate the data twin model, and update the risk model. The feedback data includes the reported data from caregivers and the data recorded by sensors. The digital twin model iteration includes health score weight iteration and environmental path model calibration.

[0051] In a third aspect, in order to solve the above technical problems, the present invention provides a nursing home priority service response device, comprising:

[0052] at least one processor;

[0053] At least one memory stores computer program code, which, when executed by the processor, causes the apparatus to perform the method of claim 1.

[0054] The above technical solution of the present invention has the following beneficial effects compared with the prior art:

[0055] The digital twin-based nursing home priority service response method, device and medium described in the present invention can capture changes in the health risks of the elderly in real time, and by constructing a lightweight digital twin model, it realizes dynamic mapping of the elderly's key health indicators and the real-time status of caregivers. By calculating the service response priority, dynamically generating a service queue, and quickly dispatching caregivers for processing, it ensures priority response during high-risk times and avoids resource mismatches and response delays caused by traditional fixed priority rules. At the same time, by collecting service execution data in real time, continuously updating the parameters in the digital twin model, and optimizing the scheduling strategy, it significantly improves the service response efficiency and resource scheduling accuracy of nursing homes in complex scenarios. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] In order to make the content of the present invention more clearly understood, the present invention is further described in detail below based on specific embodiments of the present invention in conjunction with the accompanying drawings, wherein

[0057] Figure 1 This is a flowchart of a nursing home priority service response method based on digital twins in a preferred embodiment of the present invention;

[0058] Figure 2 This is a structural block diagram of the digital twin-based nursing home service response system in the present invention. DETAILED DESCRIPTION

[0059] The present invention will be further described below with reference to the accompanying drawings and specific embodiments so that those skilled in the art can better understand the present invention and implement it. However, the embodiments are not intended to limit the present invention.

[0060] Reference Figure 1 As shown, the present invention provides a nursing home priority service response method based on digital twins, comprising the following steps:

[0061] S1: Collect health data of the elderly and status data of caregivers, and pre-process the data;

[0062] S2: Based on the preprocessed data, a dynamic model of the elderly’s health status and a dynamic model of nursing resources are constructed respectively. At the same time, an environmental path model is constructed based on the building layout data of the nursing home.

[0063] S3: Coupling the data of the elderly health status dynamic model, the nursing resource dynamic model, and the environmental path model to form a digital twin model;

[0064] S4: Calculate service response priority based on the health score, waiting time decay, and medical history characteristics of the elderly in the digital twin model;

[0065] S5: dispatching matching nursing staff to perform service response according to the service response priority;

[0066] S6: Collect the service execution results after the caregiver completes the task, and based on the service execution results, update the health score, caregiver status and service response priority in the digital twin model, and dynamically adjust the scheduling strategy.

[0067] In step S2, the dynamic model of elderly health status is used to fuse multi-source data into a single health score, which is expressed as follows:

[0068] ;

[0069] represents the health score of the elderly at time t, To score abnormal heart rate, Score abnormal blood pressure. Score the fall sign, is the risk model;

[0070] Among them, the risk model The expression is as follows:

[0071] ;

[0072] A is the age coefficient, C is the weight of chronic diseases, E is the recent abnormal events, and D is the weight of new diseases.

[0073] The above weight distribution is based on clinical priority: abnormal heart rate (40%) is directly related to the risk of sudden cardiac death, abnormal blood pressure (30%) indicates the possibility of stroke, and fall markers (20%) require timely intervention to avoid secondary injuries. (10%) Provide risk protection for patients with chronic diseases.

[0074] For example, when an elderly person suddenly has an abnormal heart rate and hour , triggering a medium-level alarm.

[0075] It should be noted that Regular recalculation is required to incorporate newly diagnosed chronic diseases or recent abnormal events to ensure the accuracy of long-term risk assessment.

[0076] In practical application, health scoring , , , The weight can be iterated according to actual needs; the iteration of weight is usually based on feedback data, such as when falls occur frequently, it needs to be improved. Weight.

[0077] The integration method of weight iteration is as follows:

[0078] Health Score The original formula is:

[0079] ;

[0080] If you need to dynamically adjust the weight of a certain item (such as the fall sign ), weight normalization must be followed so that the sum of the total sub-item weights is guaranteed to be 1 after adjustment.

[0081] If the number of falls increases significantly in the near future, weight iteration will be triggered:

[0082] Original weights: ;

[0083] After adjustment: ;

[0084] Other weights need to be reduced proportionally:

[0085] Heart rate The weight is adjusted to ;

[0086] blood pressure The weight is adjusted to ;

[0087] Basis risk The weight is kept at 0.1 (or adjusted proportionally).

[0088] The adjusted formula is: ;

[0089] In step S2 of this embodiment, the dynamic nursing resource model is used to update the location, skill label, and task queue of the nursing staff in real time. The expression is as follows:

[0090] ;

[0091] in, The location information of the caregiver, Skill labels for nursing staff; It is a task queue used to limit the maximum number of concurrent tasks; It is the working status parameter.

[0092] Specifically, the location information of the caregiver The expression is as follows:

[0093]

[0094] in, It is a coordinate system based on UWB and Bluetooth Mesh technology for positioning updates. The interval for positioning updates is less than or equal to 0.5 seconds. It is used when switching floors. To determine the floor location, is the floor calculated based on the x and y coordinates.

[0095] Caregiver location updates are a crucial foundation for the dynamic nursing resource model. This model utilizes a three-dimensional coordinate system centered at the main entrance (coordinates (0, 0, 1)), with the Z axis representing floors and underground floors represented by negative values. Location information is updated twice per second to ensure real-time and accurate data. Floor identifiers are clearly labeled to enable the system to accurately identify caregivers' locations. When moving between floors, the system considers the fixed elevator travel time cost, which defaults to 20 seconds. However, a baseline calculation is performed using a weighted average of historical wait times and dynamically calibrated based on actual caregiver travel data. The calibration formula utilizes elevator entry and exit timestamps recorded by the caregiver's UWB tag, as well as operational status information from the elevator's IoT sensors (e.g., floors stopped, door openings and closings). Furthermore, optimizations are implemented for special scenarios: during peak hours, if three consecutive elevator wait times exceed 30 seconds, the system temporarily increases the elevator travel time cost to 25 seconds. During elevator maintenance, the system automatically switches to stair travel and recalculates the travel time cost based on walking speed (e.g., stair travel time = number of floors × 10 seconds). Through this high-precision, high-frequency location update mechanism, combined with dynamic calibration and special scenario processing, the system is able to monitor the location of caregivers in real time, optimize task scheduling, and respond quickly in emergency situations.

[0096] Specifically, the expression of the skill label is as follows: ;

[0097] in, Indicates time t The skill tag collection when To represent the skill identifier; Indicates the expiration time of the skill tag; Indicates the weight of the skill label; To indicate the certificate associated with the skill tag;

[0098] Indicates that only when the certificate When the verification is true, Only included in the skill tag set middle.

[0099] Skill validity period must meet If the current time exceeds the expiration time of the skill tag, the skill tag is no longer valid, that is, it is no longer included in the skill tag set.

[0100] ;

[0101] Weight The expression is as follows:

[0102]

[0103] Skill tag management is a key component of the dynamic nursing resource model for optimizing task allocation. Each nurse is equipped with a skill tag that clearly identifies the skills the nurse possesses, such as "intravenous injection" or "first aid resuscitation." Skill tags not only help the system accurately match nurses' expertise when assigning tasks, but also dynamically adjust based on actual conditions. When a nurse's skills become invalid or no longer applicable, the system automatically removes them from the skill tag library to ensure the timeliness and accuracy of the skill tags. This dynamic management mechanism enables the system to flexibly assign tasks based on the actual capabilities of the nurses, improving the professionalism and efficiency of nursing work.

[0104] Specifically, the task queue ;

[0105] in, is the maximum number of concurrent tasks; Unique identifier for the task (e.g. "take-medication-203")

[0106] is the dynamic priority, and the calculation formula is: ;

[0107] Urgency (0 for routine / 1 for urgent); is the attenuation coefficient;

[0108] Waiting time for the task;

[0109] is a timestamp triple;

[0110] {routine care, emergency treatment, drug distribution}: task type tag;

[0111] Dynamic task queue adjustment is a core function of the dynamic nursing resource model, designed to optimize task allocation and execution efficiency. The system maintains a task queue for each caregiver, containing key information such as the task's unique identifier, dynamic priority, urgency, and task type tag. Task priority is dynamically calculated based on urgency, decay coefficient, and task wait time, ensuring that urgent tasks are prioritized.

[0112] Furthermore, it also supports task insertion constraints to prevent task queue overload, and automatically upgrades regular tasks to urgent tasks when the task waiting time is too long, while triggering an alarm reminder.

[0113] Task insertion constraints: ;

[0114] Among them, the AcceptTask function is used to decide whether to accept a new task; Indicates the j The remaining processing time of unfinished tasks, It is a threshold that represents the maximum acceptable total processing time of a task at any given point in time.

[0115] Specifically, the working state parameters ;

[0116] Device battery status (0 = exhausted, 1 = fully charged);

[0117] Work intensity index (0 = idle, 1 = overloaded);

[0118] Status mark;

[0119] The power forecast model is as follows:

[0120] ;

[0121] Parameter Description:

[0122] β: Power consumption coefficient in working state (e.g. 2mA / s for smart terminal);

[0123] γ: Charging rate (e.g. 5mA / s for wireless charging);

[0124] I(): state indicator function (value is 1 if the current state matches, otherwise 0);

[0125] Dynamic rules:

[0126] Power consumption in working state: , decreasing by 10% per second Battery capacity;

[0127] Charge compensation: When , increasing every second Battery capacity;

[0128] Low battery protection: When the power is on, it is forced to switch to charging state;

[0129] Work intensity index: ;

[0130] Parameter Description:

[0131] (10-minute sliding window); : Number of tasks completed within the window (such as taking medicine, measuring, etc.)

[0132] : Total distance moved within the window (meters); (Maximum task capacity of senior care workers)

[0133] (10-minute reasonable movement limit);

[0134] Dynamic adjustment: If , suspend assigning new tasks;

[0135] Overload protection: When , give priority to merging tasks in the same area;

[0136] State machine transition rules: state switching must be completed within 1 second; no response to task assignment during charging ( hour Lock); when or When , new tasks are temporarily suspended.

[0137] Monitoring work status parameters is a crucial component of the dynamic nursing resource model, used to assess nursing staff's work status in real time and ensure the appropriate allocation and utilization of resources. Key parameters monitored by the system include device power status, work intensity index, and status flags. Device power status is managed using a power prediction model that considers the work status power consumption coefficient, charging rate, and state indicator function to dynamically adjust power consumption and charging compensation. When the power level falls below a set threshold, the system forces the nursing staff to switch to charging mode to prevent the device from becoming unusable due to power depletion. The work intensity index is calculated within a 10-minute sliding window, taking into account the number of tasks completed and the total distance traveled within the window to assess the nursing staff's workload. When the workload exceeds a set upper limit, the system suspends the allocation of new tasks to prevent nursing staff from overwork. Furthermore, the system triggers task consolidation based on workload intensity, prioritizing tasks in the same area to reduce the workload of nursing staff. Through these monitoring and adjustment mechanisms, the system ensures that nursing staff can work efficiently at a reasonable workload while ensuring the normal operation of equipment.

[0138] In an embodiment of the present invention, the environmental path model described in step S2 is used to simulate the physical environment inside the nursing home to form a pre-calculated path time matrix; the pre-calculated path time matrix T is an N×N symmetric matrix, where N is the total number of key location nodes in the nursing home, and the matrix elements represent the estimated travel time from location node A to B, and its calculation formula is:

[0139] ;

[0140] in, is the physical distance from location A to location B; v is the walking speed of the caregiver; The estimated travel time from point A to point B; Allow for additional time when switching between elevators or stairs; It is a dynamic correction item based on real-time data.

[0141] In order to optimize the scheduling of nursing staff in this embodiment, it is necessary to dynamically update the pre-calculated path time matrix. The matrix elements of the dynamically updated pre-calculated time matrix are The expression is as follows:

[0142] ;

[0143] in, The new estimated travel time from point A to point B; is the old estimated travel time from location A to location B, that is, the estimated travel time previously obtained based on the model or historical data; is the measured travel time from location A to location B.

[0144] The path time matrix is ​​used to quantify the path length and floor switching cost (e.g., the average elevator waiting time is 20 seconds).

[0145] For example, the path time from the activity room on the 3rd floor to the ward on the 2nd floor is ;

[0146] Due to the error in the actual walking time of the caregiver (>1 minute), the matrix value is updated , to solve the path deviation caused by renovation or equipment replacement.

[0147] When an old man When there is a sudden increase in traffic, the nursing resource model screens idle nurses within a radius of 50 meters, and the path model calculates their arrival time. The data of the three are synchronized in real time through the message queue to ensure the timeliness of the decision-making basis.

[0148] In step S5, the priority of the service demand is evaluated by a multi-factor priority algorithm that combines health score, waiting time decay, and medical history characteristics. The calculation formula for the service response priority is as follows:

[0149] ;

[0150] in, For service response priority, Score your health, For the characteristics of medical history; is the waiting time decay function, is the time decay function, ;

[0151] If you have a history of myocardial infarction , directly weightedly improve priority;

[0152] The exponential decay design of the time decay function makes the impact of waiting time decrease over time, with a significant increase in the early stage and a slowdown in the later stage.

[0153] For example: (Abnormal blood pressure), wait 10 minutes, ;

[0154] Old Man B (abnormal heart rate), wait 5 minutes, ;

[0155] The system responds to Elder A first, but if Elder B's waiting time is extended to 15 minutes ( ), It will overtake, reflecting the comprehensive balance between time and risk.

[0156] In step S5 of this embodiment, the step of dispatching nursing staff to respond to the service includes:

[0157] S51. Based on the pre-calculated path time matrix, the shortest path from the current location of different caregivers to the service location is calculated for each service demand;

[0158] S52. Based on the shortest path, nursing staff skill matching, and task load, the most suitable nursing staff is dispatched for service response through a multi-objective optimization model;

[0159] S53. Identify tasks in the service response that are geographically close or can be completed consecutively in time;

[0160] S54. Evaluate the feasibility of combining tasks into a single task package.

[0161] S55. Integrate the mergeable tasks into a task package and recalculate the estimated completion time and path of the merged tasks;

[0162] S56. Update the task allocation of the nursing staff based on the merged task package.

[0163] At the nursing resource scheduling level, this system adopts the shortest path priority, skill matching priority and task batch merging mechanism. By pre-calculating the path time matrix and the skill tags stored on the blockchain, it can achieve efficient matching and task collaboration of nursing staff.

[0164] Shortest path priority principle: Based on the pre-calculated path time matrix, the travel time between floors (including 20 seconds of waiting time for the elevator) is quantified. During dynamic matching, the caregiver with the shortest path time is selected.

[0165] For example, Caregiver A is located in Area A on the 3rd floor, and it takes 61.7 seconds for Caregiver B to reach the elderly person's room; Caregiver B is in the same area on the 2nd floor, and it only takes 20 seconds → Caregiver B will be assigned first.

[0166] Skill priority rule: Skill tags are bound to blockchain evidence to ensure tamper-proofing.

[0167] The match between task requirements and nurse qualifications must be ≥80% (e.g., an infusion task requires "intravenous injection certification"). For example, if an elderly person requires CPR, the system will only screen nurses with "CPR certification," and will prioritize them even if they are slightly farther away (e.g., 30 seconds longer).

[0168] Minimum switching principle: If the caregiver's current task area overlaps with the newly required area (such as on the same floor), priority should be given to maintaining the continuity of their task.

[0169] For example, Caregiver A is handling a medication task on the 2nd floor, and a new fall alarm is issued on the same floor → Even if Caregiver B is 10 seconds closer, Caregiver A is still assigned first to reduce cross-floor movement.

[0170] In an embodiment of the present invention, the expression of the multi-objective optimization model is as follows:

[0171] ;

[0172] in, is the path time; For skill matching; For task load.

[0173] Skill Match The expression is as follows:

[0174] ;

[0175] It needs to be at least 80%.

[0176] In step S54, the specific method for evaluating the feasibility of merging tasks into one task package is as follows:

[0177] For tasks and , respectively obtain the location coordinates and the start time of the task;

[0178] When the location difference between two tasks is less than 5m and the start time difference between the two tasks is less than 10 minutes, the tasks can be merged. The expression is as follows:

[0179] ;

[0180] in, represent The location coordinates of for The location coordinates of Representative tasks of

[0181] The start time, For the task The start time of the

[0182] Priority of new tasks after merging: ;

[0183] in, is the priority of the new task after the merger; and is the priority of the two tasks to be merged; Indicates taking the larger value of the two priorities. is an indicator function, when and equal When , its value is 1, otherwise it is 0; here and Represents the properties of two tasks (such as type, category, etc.).

[0184] When multiple tasks have similar characteristics or are located in the same area, they can be merged into a new task. The priority of the merged new task will be recalculated based on the priority of the merged task. Through these dynamic adjustment mechanisms, it is possible to flexibly respond to changing work needs, optimize the task allocation of nursing staff, and improve overall work efficiency.

[0185] In this embodiment, in step S7, the service execution result includes task completion status, actual response time, task execution quality, nursing staff feedback, system records and logs, and health status updates.

[0186] In actual applications, nursing staff use mobile terminals to record nursing tasks in real time and report actual handling results (such as selecting preset options (such as "fall without fracture" and "false alarm of abnormal heart rate") or filling in free-text notes, such as supplementing execution details (such as "the elderly refuse to take medicine")), and generate a full-process traceable log; at the same time, the response time, path deviation and handling results are synchronized to the digital twin model to drive the optimization of health score weights (such as increasing the priority of high-frequency events) and path calibration, so as to realize the quantitative evaluation of service quality and continuous self-optimization of scheduling strategies.

[0187] After the caregiver completes the task, the system automatically records the actual time from dispatch to the caregiver's arrival; the difference between the caregiver's actual walking time and the estimated value (such as delays caused by temporary obstacles and detours) and other indicators.

[0188] To more clearly explain the workflow of the digital twin-based nursing home priority service response method in this embodiment, the following two specific application scenarios for elderly people are used for description:

[0189] 21:30: Elderly A, 80 years old, history of heart failure, =0.44 , in the activity room on the 3rd floor, a sudden abnormal heart rate (>120 beats / min for 3 consecutive minutes, ), the system calculates its health score:

[0190] Triggering a medium-level alarm;

[0191] 21:31: Elderly B, 75 years old, with a history of diabetes, fell in an adjacent room on the third floor (Δa = 3.5 m / s², immobility time > 30 seconds), and the system flagged the fall. , calculate its health score: ;

[0192] The system response process is as follows:

[0193] Priority calculation: Elder A's priority (waiting for 1 minute, M=1): ;

[0194] Elderly B priority (wait 0 minutes) ;

[0195] Conclusion: Give priority to responding to Elder A.

[0196] Nursing staff matching: Screen the available nurses on the 3rd floor: Nurse A (skill: cardiopulmonary resuscitation certification, task load q=1), Nurse B (skill: basic nursing, q=2).

[0197] Through multi-objective optimization model Calculate the matching score:

[0198] Nurse A: Path time T=20 seconds, skill matching is 100%, load weight is 0.2;

[0199] ;

[0200] Caregiver B: Path time T = 15 seconds, skill matching 0%, load weight is 0.4:

[0201] ;

[0202] Conclusion: Nurse A has a high skill match and is assigned first.

[0203] Batch task merging: The system detects that Elder B needs to be on the same floor as Elder A and the time interval is less than 5 minutes, and generates a combined task: After Caregiver A finishes treating Elder A, he immediately assists Elder B in taking medicine to reduce cross-room movement.

[0204] Task execution and feedback: Caregiver A arrives at Elder A's room, scans the code to verify his identity, and completes the abnormal heart rate treatment within 5 minutes.

[0205] The sensor detects that Elder A has resumed lying down and his heart rate has dropped to 90 beats per minute, and the system automatically marks the task as completed.

[0206] Nurse A went to the elderly man's room B to help him take medicine, and the time of ineffective movement was reduced by 40%.

[0207] Path calibration: Caregiver A's actual walking time is 2 minutes longer than the precalculated time (due to a temporary obstacle and a detour). Update the path matrix: Second;

[0208] Weight adjustment: The system analyzes that there is no delay in this response and maintains the weight of the fall mark .

[0209] like Figure 2 As shown, the present invention provides a nursing home priority service response system based on digital twins, including:

[0210] The perception layer includes a physiological detection module and an environmental positioning module; the physiological detection module includes a heart rate sensor, a blood pressure sensor, and an activity monitoring sensor for real-time collection of the elderly's physiological indicators; the environmental positioning module is used to monitor and record the location and trajectory of the elderly and caregivers in the nursing home;

[0211] The edge computing layer is used to preprocess data and build a digital twin model to map the elderly's health status and nursing resource status in real time; the digital twin model includes a dynamic model of the elderly's health status, a dynamic model of nursing resources, and an environmental path model;

[0212] A decision-making layer, configured to dynamically calculate service response priorities and optimize resource allocation based on nursing staff's skills, location, and task load; the decision-making layer calculates service response priorities using a multi-factor priority algorithm;

[0213] The execution layer is used to send service instructions to nursing staff;

[0214] The feedback layer is used to collect feedback data, iterate the data twin model, and update the risk model. The feedback data includes the reported data from caregivers and the data recorded by sensors. The digital twin model iteration includes health score weight iteration and environmental path model calibration.

[0215] In this embodiment, multi-sensor integration technology is used to integrate the heart rate monitoring sensor, blood pressure sensor and activity monitoring sensor into a multi-sensor integrated health monitoring system to comprehensively monitor the health status of the elderly. The multi-sensor integrated health monitoring system also includes a data processing unit, a health status assessment module and an alarm module. The data processing unit receives the data detected by the sensor and pre-processes the data to remove noise, fuse the data and monitor abnormalities. The health status assessment module evaluates the immediate and long-term monitoring status of the elderly based on the fused data. When an abnormal health indicator is detected, the alarm module is triggered to alarm and notify caregivers or medical personnel.

[0216] The environmental positioning module in this embodiment uses UWB, BLE, Wi-Fi, 5G, and wall positioning technologies for positioning, and fuses multiple different positioning results to output the optimal result;

[0217] UWB measures the signal propagation time difference (TDoA) by using the formula Calculate the distance between the device and the base station;

[0218] BLE is based on the received signal strength RSSI, through the attenuation model Estimate the relative distance between the device and the beacon; where n is the path loss factor and A is the signal strength at the reference distance;

[0219] Wi-Fi uses fingerprint matching method to minimize signal differences Achieve sub-meter positioning;

[0220] in, is the currently observed RSSI value, is the RSSI value in the database;

[0221] 5G obtains the optimal positioning position by fusing the signal propagation time difference TDoA and the angle of arrival AOA of the signal. The expression of the optimal positioning position is as follows:

[0222] ;

[0223] in, Indicates the optimal positioning position, that is, the most likely actual position obtained through optimization calculation, that is, the most likely actual position obtained through optimization calculation; is the jth time difference of the i-th measurement; c is the speed of light, which is a constant; is the jth angle of the i-th measurement;

[0224] When locating a wall, the original measured distance is adjusted using the reflection angle correction formula to obtain the corrected true distance. The reflection angle correction formula is expressed as follows: ;

[0225] is the corrected distance, i.e. the more accurate distance after adjustment; is the original distance, the uncorrected measured distance; is the correction factor, which is used to control the adjustment range of the original distance measurement value. is the reflection angle;

[0226] On the basis of wall positioning, we further utilize the deep coupling of wall-assisted positioning technology and error correction algorithm to estimate the deviation caused by signal reflection by detecting the known position, material and geometric characteristics of indoor walls. , and then use the angle correction formula to correct the measured angle to improve positioning accuracy;

[0227] The angle correction formula is as follows: ;

[0228] in, To measure angles, is the angle after correction.

[0229] A weighted data fusion strategy is used during data fusion to dynamically assign weights to each technical result. During the data preprocessing phase, a confidence assessment is performed on each positioning result, and the credibility score of each technical result is calculated based on factors such as signal strength, environmental interference, device performance, and multipath effects. , where the weight , dynamically allocated using the following formula: ;

[0230] After data fusion, a more accurate final position is output, and its expression is as follows:

[0231] ;

[0232] is the final fusion result; is the position result obtained by the i-th positioning module; is the weight of the i-th positioning module; N is the number of positioning modules.

[0233] Preferably, the data processing unit processes the data through edge computing to ensure the accuracy and timeliness of the data, providing a reliable basis for subsequent health analysis.

[0234] Specifically, the workflow of the digital twin-based nursing home service response system is as follows:

[0235] 1. At the perception layer, the physiological monitoring module and environmental positioning module are used to collect the elderly’s physiological data and the location information of caregivers in real time;

[0236] 2. Preprocess the collected data at the edge computing layer, including noise filtering and anomaly detection;

[0237] At the edge computing layer, based on lightweight digital twin models and dynamic mapping, environmental path models, dynamic nursing resource models, and elderly health scores are established;

[0238] 3. Use a multi-factor prioritization algorithm at the decision-making level to integrate health scores, wait time decay, and medical history characteristics to assess the priority of service needs;

[0239] At the decision-making level, the optimal nursing staff matching plan is determined based on the shortest path priority principle, skill priority rule, minimum switching principle and multi-objective planning;

[0240] 4. Spatial clustering of multiple requirements in the same area at the decision-making level, merging them into a single task package;

[0241] 5. Send instructions to nursing staff at the execution layer, support code scanning verification, task progress reporting and path navigation;

[0242] 6. Collect task execution results at the feedback layer, including data reported by caregivers and data recorded by sensors; iterate the digital twin model based on the collected data at the feedback layer, including health score weight iteration, risk model update, and path model calibration.

[0243] By building a lightweight digital twin model, the system dynamically maps key health indicators (heart rate, blood pressure, and abnormal activity) of elderly individuals with the real-time status of caregivers (location, skills, and workload). This effectively reduces the computing power requirements of traditional digital twins, making them suitable for actual nursing home deployment environments. Incorporating multimodal data fusion technology, the system captures changes in elderly individuals' health risks in real time. It dynamically calculates service queues using a multi-factor priority algorithm, comprehensively considering health scores, wait time decay effects, and medical history characteristics. This ensures that high-risk events are prioritized, avoiding the resource mismatch and response delays caused by traditional fixed priority rules.

[0244] At the nursing resource scheduling level, the system adopts the shortest path priority, skill matching priority and task batch merging mechanism, and realizes efficient matching of nursing staff and task collaboration by pre-calculating the path time matrix and storing skill labels on the blockchain.

[0245] The closed-loop feedback mechanism collects service execution data in real time, continuously calibrates health score weights, optimizes path models, and updates long-term risk assessment parameters, making the system adaptive and able to dynamically respond to environmental changes and demand fluctuations.

[0246] At the same time, through the digital traceability and quality assessment mechanism, human scheduling errors are significantly reduced, providing a reliable solution for nursing homes to achieve precise and intelligent services under limited human resources.

[0247] The invention provides a nursing home priority service response device, comprising:

[0248] at least one processor;

[0249] At least one memory stores computer program code, which, when executed by the processor, causes the apparatus to perform the method of claim 1.

[0250] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware. Furthermore, the present application may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0251] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0252] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0253] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1 The steps for the function specified in one or more boxes.

[0254] Obviously, the above embodiments are merely examples for clarity of explanation and are not intended to limit the implementation methods. Those skilled in the art will appreciate that other variations or modifications can be made based on the above description. It is not necessary and impossible to enumerate all implementation methods here. Obvious variations or modifications arising therefrom remain within the scope of protection of the present invention.

Claims

1. A digital twin-based nursing home priority service response method, characterized by: The following steps are involved: S1: Collect health data of the elderly and status data of caregivers, and pre-process the data; S2: Based on the preprocessed data, a dynamic model of the elderly’s health status and a dynamic model of nursing resources are constructed respectively. At the same time, an environmental path model is constructed based on the building layout data of the nursing home. S3: Coupling the data of the elderly health status dynamic model, the nursing resource dynamic model, and the environmental path model to form a digital twin model; S4: Calculate service response priority based on the health score, waiting time decay, and medical history characteristics of the elderly in the digital twin model; S5: dispatching matching nursing staff to perform service response according to the service response priority; S6: Collect the service execution results after the caregiver completes the task, and based on the service execution results, update the health score, caregiver status and service response priority in the digital twin model, and dynamically adjust the scheduling strategy.

2. The digital twin-based nursing home priority service response method according to claim 1 is characterized in that: In step S2, the dynamic model of elderly health status is used to fuse multi-source data into a single health score, which is expressed as follows: ; represents the health score of the elderly at time t, To score abnormal heart rate, Score abnormal blood pressure. Score the fall sign, is the risk model; Among them, the risk model The expression is as follows: ; A is the age coefficient, C is the weight of chronic diseases, E is the recent abnormal events, and D is the weight of new diseases.

3. The digital twin-based nursing home priority service response method according to claim 1 is characterized in that: In step S2, the dynamic nursing resource model is used to update the nursing staff's location, skill tags, and task queue in real time. Its expression is as follows: ; in, The location information of the caregiver, Skill labels for nursing staff; It is a task queue used to limit the maximum number of concurrent tasks; It is the working status parameter.

4. The digital twin-based nursing home priority service response method according to claim 1 is characterized in that: The environmental path model in step S2 is used to simulate the physical environment inside the nursing home to form a pre-calculated path time matrix; the pre-calculated path time matrix T is an N×N symmetric matrix, where N is the total number of key location nodes in the nursing home, and the matrix elements are It represents the estimated travel time from location node A to B, and its calculation formula is: ; in, is the physical distance from location A to location B; v is the walking speed of the caregiver; The estimated travel time from point A to point B; Allow for additional time when switching between elevators or stairs; It is a dynamic correction item based on real-time data.

5. The digital twin-based nursing home priority service response method according to claim 4 is characterized in that: In step S4, the priority of the service demand is evaluated by a multi-factor priority algorithm that combines health score, waiting time decay, and medical history characteristics. The calculation formula for the service response priority is as follows: ； Among them, P i For service response priority, is the health score, M is the medical history characteristics; is the waiting time decay function, .

6. The digital twin-based nursing home priority service response method according to claim 4 is characterized in that: In step S5, the step of dispatching nursing staff to respond to the service includes: S51: Based on the pre-calculated path time matrix, calculate the shortest path from the current location of different caregivers to the service location for each service demand; S52: Based on the shortest path, nursing staff skill matching and task load, the most suitable nursing staff is dispatched for service response through a multi-objective optimization model; S53: Identify tasks in the service response that are geographically close or can be completed consecutively in time; S54: Evaluate the feasibility of merging tasks into a task package; S55: Integrate the mergeable tasks into a task package, and recalculate the estimated completion time and path of the merged tasks; S56: Update the task allocation of the nursing staff according to the merged task package.

7. The digital twin-based nursing home priority service response method according to claim 6 is characterized in that: The expression of the multi-objective optimization model is as follows: ； in, is the path time; For skill matching; For task load.

8. The digital twin-based nursing home priority service response method according to claim 1 is characterized in that: In step S6, the service execution results include task completion status, actual response time, task execution quality, nursing staff feedback, system records and logs, and health status updates.

9. A digital twin-based nursing home priority service response system, characterized by: include: The perception layer includes a physiological detection module and an environmental positioning module; the physiological detection module includes a heart rate sensor, a blood pressure sensor, and an activity monitoring sensor for real-time collection of the elderly's physiological indicators; the environmental positioning module is used to monitor and record the location and trajectory of the elderly and caregivers in the nursing home; The edge computing layer is used to preprocess data and build a digital twin model to map the elderly's health status and nursing resource status in real time; the digital twin model includes a dynamic model of the elderly's health status, a dynamic model of nursing resources, and an environmental path model; A decision-making layer, configured to dynamically calculate service response priorities and optimize resource allocation based on nursing staff's skills, location, and task load; the decision-making layer calculates service response priorities using a multi-factor priority algorithm; The execution layer is used to send service instructions to nursing staff; The feedback layer is used to collect feedback data, iterate the data twin model, and update the risk model. The feedback data includes the reported data from caregivers and the data recorded by sensors. The digital twin model iteration includes health score weight iteration and environmental path model calibration.

10. A nursing home priority service response device, characterized by: include: at least one processor; At least one memory storing computer program code, which, when executed by a processor, causes the apparatus to perform the method according to any one of claims 1 to 8.

Images