Intelligent nursing environment monitoring and infection control system

Abstract

The invention relates to the technical field of medical care, in particular to an intelligent nursing environment monitoring and infection control system which comprises a data acquisition module, a data transmission module, a data processing and analysis module, a disinfection effect evaluation module, a control execution module, a user interaction module and an early warning response module. According to the scheme, the nursing environment is monitored by adopting the sensor network and the video monitoring equipment; a multi-agent-based communication cooperative control method is adopted to automatically monitor and control a nursing environment, and various operations from patient monitoring, medicine distribution to illness state reporting and the like are ensured to be executed orderly by performing topological sorting on various tasks in a nursing process; performing time-space conversion on the original binary communication graph to optimize the communication efficiency between the monitoring devices in the nursing environment; and through the design of a control rule, the data change is responded by adopting an event-driven mode, so that the influence of time delay and data packet loss is reduced.

Description

Technical Field

[0001] The present invention relates to the field of medical care technology, and in particular to an intelligent care environment monitoring and infection control system. Background Art

[0002] At present, in medical care institutions, in order to ensure the health and safety of patients and prevent the occurrence of hospital-acquired infections, the nursing environment needs to be strictly monitored. Traditional detection methods often rely on manual operations, which are not only inefficient but also prone to omissions or errors. However, with the advancement of Internet of Things technology, more and more environmental monitoring methods have emerged. Despite this, these monitoring methods have not been able to effectively address the challenges of infection control in the nursing environment due to the lack of integrated management and insufficient intelligence. In addition, the existing system fails to achieve intelligent regulation of equipment in the nursing environment, and lacks an architecture that can automatically monitor and improve environmental conditions. At the same time, there are still slow responses and data packet losses in the currently used control systems. These problems lead to the untimely control of control measures, which in turn has an adverse impact on the quality of the nursing environment. Summary of the invention

[0003] In view of the above situation, in order to overcome the defects of the prior art, the present invention provides an intelligent nursing environment monitoring and infection control system. In view of the problem that the existing system fails to effectively deal with the infection control in the nursing environment due to the lack of integrated management and insufficient intelligence level, the nursing environment is monitored by using a sensor network and a video monitoring device; in view of the problem that the existing system fails to realize the intelligent regulation of the equipment in the nursing environment and lacks the architecture that can automatically monitor and improve the environmental conditions, a multi-agent-based communication collaborative control method is used to automatically monitor and control the nursing environment, and a communication network model containing a set of agents of all monitoring devices is defined. By topologically sorting the tasks in the nursing process, the orderly execution of various operations such as patient monitoring, drug delivery and disease reporting is ensured, and task conflicts or omissions are avoided; at the same time, by performing time-space conversion on the original binary communication graph, a relaxation graph with continuous values ​​is formed, so as to optimize the communication efficiency between the monitoring devices in the nursing environment; in view of the problem that the control system currently used has slow response and data packet loss, resulting in untimely control measures, the multi-agent-based communication collaborative control method uses an event-driven method to respond to data changes through the design of control rules, thereby reducing the impact of time delays and data packet loss.

[0004] The intelligent nursing environment monitoring and infection control system provided by the present invention includes a data acquisition module, a data transmission module, a data processing and analysis module, a disinfection effect evaluation module, a control execution module, a user interaction module and an early warning response module;

[0005] The data acquisition module uses a sensor network and video monitoring equipment to monitor and collect data on the nursing environment 24 hours a day to obtain nursing environment data;

[0006] The data transmission module uses wireless technology and Ethernet to transmit the collected nursing environment data to the central server of the data processing and analysis module for processing;

[0007] The data processing and analysis module uses a central server to store and process the nursing environment data, and uses a multi-agent-based communication and collaborative control method to perform data analysis and prediction to obtain nursing environment prediction results and control strategies;

[0008] The disinfection effect evaluation module obtains the hospital environment nursing standards, which are used to evaluate the disinfection effects of wards, operating rooms and infection nursing areas in the nursing environment. According to the nursing environment data and hospital environment nursing standards obtained by the data acquisition module, a scoring system from 1 to 10 is used to evaluate the disinfection effects of different areas, where 1 indicates a very poor disinfection effect and 10 indicates an excellent disinfection effect, to obtain a disinfection effect score;

[0009] The control execution module automatically adjusts the working state of each device in the nursing environment according to the obtained control strategy, and detects it through the disinfection effect evaluation module to obtain a disinfection effect score. According to the disinfection effect score, the control strategy is obtained through the data processing and analysis module, and control adjustments are made to ensure that the cleaning standard is met and a suitable indoor environment is maintained;

[0010] The user interaction module provides a graphical user interface to display monitored care environment data and system status information;

[0011] The early warning response module sets a safety range for the nursing environment data, and issues an alarm in a timely manner when the nursing environment data exceeds the safety range. The user interaction module also notifies medical staff via SMS, email, or APP reminders so that quick action can be taken.

[0012] Furthermore, the data processing and analysis module uses a multi-agent-based communication and collaborative control method to perform data analysis and prediction. The multi-agent-based communication and collaborative control method specifically includes the following steps:

[0013] Step S1: Define the communication network model, take the devices in the sensor network as agents, and represent the communication relationship between agents as a binary communication graph ,in, is the set of intelligent agents, is the edge set, is the adjacency matrix;

[0014] Step S2: Multi-agent communication: According to the communication network model, the agents process tasks and exchange information in order according to the topological sorting to obtain a solution;

[0015] Step S3: Spatiotemporal graph conversion, converting the original binary communication graph Convert to a relaxation graph with continuous values , the formula used is as follows: ; ;

[0016] In the formula, Represents a binary communication graph The adjacency matrix of and denote the spatial and temporal adjacency matrices respectively, and represents a differentiable graph mask, and denote the spatial and temporal relaxation graphs respectively, Represents the relaxation graph The adjacency matrix of

[0017] Step S4: Optimize the relaxation graph spatiotemporal connectivity;

[0018] Step S5: Control rule construction, using controllers, buffers and actuators to respond to data in an event-driven manner, perform automated control, and control rule design. The formula used is as follows: ;

[0019] In the formula, is the controller gain matrix, and is the index, Representing an Agent The neighbor set of represents the elements of the adjacency matrix, represents the sampling time, represents the time interval for sampling the agent state, and Represents the states of different agents, Representing an Agent In time Control input of

[0020] Step S6: Define time delay , the formula used is as follows: ;

[0021] In the formula, Indicates time delay, Indicates time, t∈[kh + t k , (k+1) h + t k +1 ] , and Respectively represent Second and Total delay of sub-sampling instants;

[0022] Step S7: Set packet loss time limit, set limits on the frequency and duration of packet loss, define the time interval and , Indicates that the time delay is [ d 1 , d 2 ] The total time length in the interval, Indicated in [ d 2 , d 3 ] The total time length of the interval, where , Indicates the minimum delay, and Respectively and The upper bound of the time delay.

[0023] Furthermore, in step S2, multi-agent communication specifically includes the following steps:

[0024] Step S21: Initialization, setting the initial state and role of the agent, and determining the plug-in configuration of each agent;

[0025] Step S22: query distribution, given a task q, distribute task q to all agents in sequence;

[0026] Step S23: Topological sorting, sorting the dependencies between agents to obtain a directed acyclic spatial communication graph ;

[0027] Step S24: message passing, each agent generates a response based on the task q, the currently set role, the state, and the information of its temporal neighbors and spatial neighbors. Temporal neighbors refer to the agents that have interacted with it in the previous round of dialogue, and spatial neighbors refer to the agents that have interacted with it in the same round of dialogue;

[0028] Step S25: Iterate and update the summary, repeating steps S23 and S24 until M rounds of dialogue are completed and a final solution is generated;

[0029] Further, in step S4, the relaxation graph is optimized The spatiotemporal connectivity of

[0030] Step S41: Distribution approximation, using the policy gradient method to reflect the relaxation graph The underlying probability distribution of

[0031] Step S42: Low rank sparsity, minimize the nuclear norm of the graph mask, obtain the optimized graph mask, and remove redundant information. The formula used is as follows: ;

[0032] in, Represents looking for a set of parameters , get the minimum value, is the set of all image masks, and Representation mask and The elements in For a collection Each element in Perform a sum operation, is the nuclear norm, is the constraint character, is the original matrix, is the optimized image mask, is the constraint condition of the Frobenius norm, is the threshold value set;

[0033] Step S43: Perform a pruning operation on the optimized graph mask to remove redundant connections and obtain a binary mask matrix. The formula used is as follows: ;

[0034] In the formula, is the characteristic function, is a binary mask matrix, represents the total number of elements in the adjacency matrix, represents the pruning ratio, Before selection elements, is a logical operator;

[0035] Step S44: Update connectivity by applying the binary mask matrix to the adjacency matrix , we get the updated connectivity using the following formula: ;

[0036] In the formula, and represents the spatial and temporal binary mask matrices, Represents a binary communication graph The subgraph of Represents the adjacency matrix of the subgraph after pruning optimization.

[0037] The beneficial effects achieved by the present invention using the above scheme are as follows:

[0038] (1) In view of the fact that the lack of integrated management and insufficient intelligence level have prevented the effective response to infection control issues in the nursing environment, sensor networks and video surveillance equipment are used to monitor the nursing environment. The sensor network monitors various parameters in the nursing environment in real time, and uses video analysis technology to identify potential sources of infection or improper operating behaviors. In view of the fact that the existing models fail to achieve intelligent regulation of equipment in the nursing environment and lack an architecture that can automatically monitor and improve environmental conditions, a multi-agent-based communication collaborative control method is used to automatically monitor and control the nursing environment. The system defines a communication network model that includes a collection of agents for all monitoring devices. By topologically sorting the tasks in the nursing process, it ensures the orderly execution of various operations such as patient monitoring, drug delivery, and disease reporting, avoiding task conflicts or omissions. At the same time, by performing time-space conversion on the original binary communication graph, a relaxation graph with continuous values ​​is formed, thereby optimizing the communication efficiency between monitoring devices in the nursing environment and ensuring the efficient operation of the entire system.

[0039] (2) In view of the problem that the current control system has slow response and data packet loss, which leads to the untimely control measures and thus has an adverse impact on the quality of the nursing environment, a network architecture that can effectively reduce time delays and data packet loss is designed based on the communication and collaborative control method of multiple agents. Through the design of control rules, an event-driven approach is adopted to respond to data changes, and combined with the precise definition and management of time delays and data packet loss phenomena, the response speed and stability of the system are significantly improved, thereby improving the overall quality of the nursing environment. Even in complex or unpredictable situations, the system can ensure the timely and accurate execution of various control measures, improve the efficiency of nursing work, and reduce the difficulty of improving the nursing environment. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] Figure 1 A schematic diagram of the intelligent nursing environment monitoring and infection control system provided by the present invention;

[0041] The accompanying drawings are used to provide further understanding of the present invention and constitute a part of the specification. They are used to explain the present invention together with the embodiments of the present invention and do not constitute a limitation of the present invention. DETAILED DESCRIPTION

[0042] The technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments; based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0043] Example 1, see Figure 1 , the intelligent nursing environment monitoring and infection control system provided by the present invention includes a data acquisition module, a data transmission module, a data processing and analysis module, a disinfection effect evaluation module, a control execution module, a user interaction module and an early warning response module;

[0044] The data acquisition module uses a sensor network and video monitoring equipment to monitor and collect data on the nursing environment 24 hours a day to obtain nursing environment data;

[0045] The data transmission module uses wireless technology and Ethernet to transmit the collected nursing environment data to the central server of the data processing and analysis module for processing;

[0046] The data processing and analysis module uses a central server to store and process the nursing environment data, and uses a multi-agent-based communication and collaborative control method to perform data analysis and prediction to obtain nursing environment prediction results and control strategies;

[0047] The disinfection effect evaluation module obtains the hospital environment nursing standards, which are used to evaluate the disinfection effects of wards, operating rooms and infection nursing areas in the nursing environment. According to the nursing environment data and hospital environment nursing standards obtained by the data acquisition module, a scoring system from 1 to 10 is used to evaluate the disinfection effects of different areas, where 1 indicates a very poor disinfection effect and 10 indicates an excellent disinfection effect, to obtain a disinfection effect score;

[0048] The control execution module automatically adjusts the working state of each device in the nursing environment according to the obtained control strategy, and detects it through the disinfection effect evaluation module to obtain a disinfection effect score. According to the disinfection effect score, the control strategy is obtained through the data processing and analysis module, and control adjustments are made to ensure that the cleaning standard is met and a suitable indoor environment is maintained;

[0049] The user interaction module provides a graphical user interface to display monitored care environment data and system status information;

[0050] The early warning response module sets a safety range for the nursing environment data, and issues an alarm in a timely manner when the nursing environment data exceeds the safety range. The user interaction module also notifies medical staff via SMS, email, or APP reminders so that quick action can be taken.

[0051] Embodiment 2: This embodiment is based on the above embodiment. In the data acquisition module, the sensor network includes sensor devices for detecting temperature, humidity, air quality, light, sound, activity detection, and disinfection status. In a specific implementation, a DHT11 / DHT22 temperature and humidity sensor is installed in the nursing environment to monitor indoor temperature and humidity, a Plantower PMS series sensor measures particulate matter concentration, an MH-Z19 CO2 sensor detects carbon dioxide concentration, a BH1750FVI digital light intensity sensor measures light intensity, a GM100 sound sensor detects environmental noise, an HC-SR501 infrared human body sensing module senses human infrared radiation, detects whether there is human activity in the nursing environment area, and an ultraviolet intensity meter measures the radiation intensity of the ultraviolet disinfection lamp.

[0052] The disinfection effect evaluation module evaluates the disinfection effect of the current nursing environment based on the nursing environment data and hospital environmental nursing standards obtained by the data acquisition module. Specifically, it uses the nursing environment data of temperature, humidity, air quality, light, sound, activity detection and disinfection status monitored by the sensor network. The evaluation content covers aspects such as surface cleanliness, air quality and medical equipment cleaning, and gives corresponding disinfection effect scores according to the hospital's hygiene standards and specific requirements of different areas, and finally forms a comprehensive score. A score of 6 is basically qualified;

[0053] For example, a surface cleanliness score of 9 means that after strict cleaning and disinfection, no harmful microorganisms were detected on the surfaces of all objects; a surface cleanliness score of 7 means that most surfaces are clean, but a small amount of bacteria were found in some corners; a surface cleanliness score of 5 means that drug-resistant strains were found at contact points in high-risk areas, and the cleanliness does not meet the standards and needs to be re-cleaned;

[0054] At the same time, according to the instructions of the control execution module, the equipment for improving the nursing environment is intelligently controlled in the sensor network. The equipment for improving the nursing environment includes air conditioning, ventilation systems, air purifiers, humidifiers, dehumidifiers, lighting systems, disinfection equipment and noise reduction devices. These devices improve the detected temperature, humidity, air quality, etc. to provide a good nursing environment.

[0055] Embodiment three, this embodiment is based on the above embodiment, the user interaction module provides a graphical user interface, the graphical user interface includes an environment overview page, which uses a bar graph to display the temperature fluctuations in the past week, and a line graph to display the humidity change trend; a real-time monitoring panel, which displays the status of all current monitoring points and highlights abnormal conditions. When the value of any monitoring point exceeds a preset threshold, the corresponding icon flashes and displays the specific value; operation log recording, an operation log database is established in the background of this system to record the details of each login, parameter adjustment, equipment control and other operations; a help center is opened with built-in video tutorials, graphic guides, and online customer service support.

[0056] Embodiment 4, based on the above embodiment, the data processing and analysis module uses a multi-agent based communication and collaborative control method to perform data analysis and prediction. The multi-agent based communication and collaborative control method specifically includes the following steps:

[0057] Step S1: Define the communication network model, take the devices in the sensor network as agents, and represent the communication relationship between agents as a binary communication graph ,in, is the set of intelligent agents, is the edge set, is the adjacency matrix;

[0058] Step S2: Multi-agent communication: According to the communication network model, the agents process tasks and exchange information in order according to the topological sorting to obtain a solution;

[0059] Step S3: Spatiotemporal graph conversion, converting the original binary communication graph Convert to a relaxation graph with continuous values , the formula used is as follows: ; ;

[0060] In the formula, Represents a binary communication graph The adjacency matrix of and denote the spatial and temporal adjacency matrices respectively, and represents a differentiable graph mask, and denote the spatial and temporal relaxation graphs respectively, Represents the relaxation graph The adjacency matrix of

[0061] Step S4: Optimize the relaxation graph spatiotemporal connectivity;

[0062] Step S5: Control rule construction, using controllers, buffers and actuators to respond to data in an event-driven manner, perform automated control, and control rule design. The formula used is as follows: ;

[0063] In the formula, is the controller gain matrix, and is the index, Representing an Agent The neighbor set of represents the elements of the adjacency matrix, represents the sampling time, represents the time interval for sampling the agent state, and Represents the states of different agents, Representing an Agent In time Control input of

[0064] Step S6: Define time delay , the formula used is as follows: ;

[0065] In the formula, Indicates time delay, Indicates time, t∈[kh + t k , (k+1) h + t k +1 ] , and Respectively represent Second and Total delay of sub-sampling instants;

[0066] Step S7: Set packet loss time limit, set limits on the frequency and duration of packet loss, define the time interval and , Indicates that the time delay is [ d 1 , d 2 ] The total time length in the interval, Indicated in [ d 2 , d 3 ] The total time length of the interval, where , Indicates the minimum delay, and Respectively and The upper bound of the time delay.

[0067] Embodiment 5, based on the above embodiment, in step S2, multi-agent communication specifically includes the following steps:

[0068] Step S21: Initialization, setting the initial state and role of the agent, and determining the plug-in configuration of each agent;

[0069] Step S22: query distribution, given a task q, distribute task q to all agents in sequence;

[0070] Step S23: Topological sorting, sorting the dependencies between agents to obtain a directed acyclic spatial communication graph ;

[0071] Step S24: message passing, each agent generates a response based on the task q, the currently set role, the state, and the information of its temporal neighbors and spatial neighbors. Temporal neighbors refer to the agents that have interacted with it in the previous round of dialogue, and spatial neighbors refer to the agents that have interacted with it in the same round of dialogue;

[0072] Step S25: Iterate and update the summary, repeating steps S23 and S24 until M rounds of dialogue are completed and a final solution is generated.

[0073] By performing the above operations, in order to effectively deal with the problem of infection control in the nursing environment due to the lack of integrated management and insufficient intelligence level, sensor networks and video monitoring equipment are used to monitor the nursing environment. The sensor network monitors various parameters in the nursing environment in real time, and identifies potential sources of infection or improper operation behaviors through video analysis technology; in order to solve the problem that the existing model fails to realize the intelligent regulation of equipment in the nursing environment and lacks the architecture that can automatically monitor and improve the environmental conditions, a multi-agent-based communication collaborative control method is used to automatically monitor and control the nursing environment. The system defines a communication network model containing a collection of agents for all monitoring devices. By topologically sorting the tasks in the nursing process, the orderly execution of various operations such as patient monitoring, drug delivery and disease reporting is ensured, avoiding task conflicts or omissions; at the same time, by performing time-space conversion on the original binary communication graph, a relaxation graph with continuous values ​​is formed, so as to optimize the communication efficiency between monitoring devices in the nursing environment and ensure the efficient operation of the entire system;

[0074] Embodiment 6: This embodiment is based on the above embodiment. In step S4, the relaxation graph is optimized. The spatiotemporal connectivity of

[0075] Step S41: Distribution approximation, using the policy gradient method to reflect the relaxation graph The underlying probability distribution of

[0076] Step S42: Low rank sparsity, minimize the nuclear norm of the graph mask, obtain the optimized graph mask, and remove redundant information. The formula used is as follows: ;

[0077] in, Represents looking for a set of parameters , get the minimum value, is the set of all image masks, and Representation mask and The elements in For a collection Each element in Perform a sum operation, is the nuclear norm, is the constraint character, is the original matrix, is the optimized image mask, is the constraint condition of the Frobenius norm, is the threshold value set;

[0078] Step S43: Perform a pruning operation on the optimized graph mask to remove redundant connections and obtain a binary mask matrix. The formula used is as follows: ;

[0079] In the formula, is the characteristic function, is a binary mask matrix, represents the total number of elements in the adjacency matrix, represents the pruning ratio, Before selection elements, is a logical operator;

[0080] Step S44: Update connectivity by applying the binary mask matrix to the adjacency matrix , we get the updated connectivity using the following formula: ;

[0081] In the formula, and represents the spatial and temporal binary mask matrices, Represents a binary communication graph The subgraph of Represents the adjacency matrix of the subgraph after pruning optimization.

[0082] By performing the above operations, a network architecture that can effectively reduce time delays and packet loss is designed based on the communication and collaborative control method of multiple agents to address the problem of slow response and packet loss in the currently used control system, which leads to untimely control measures and thus adversely affects the quality of the nursing environment. Through the design of control rules, an event-driven approach is adopted to respond to data changes, combined with the precise definition and management of time delays and packet loss, the response speed and stability of the system are significantly improved, thereby improving the overall quality of the nursing environment. Even in complex or unpredictable situations, the system can ensure the timely and accurate execution of various control measures, improve the efficiency of nursing work, and reduce the difficulty of improving the nursing environment.

[0083] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device.

[0084] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

[0085] The present invention and its embodiments are described above, and such description is not restrictive. The drawings show only one embodiment of the present invention, and the actual structure is not limited thereto. In short, if ordinary technicians in the field are inspired by it, without departing from the purpose of the invention, they can design a structure and embodiment similar to the technical solution without creativity, which should belong to the protection scope of the present invention.

Claims

1. Intelligent nursing environment monitoring and infection control system, characterized by: It includes data acquisition module, data transmission module, data processing and analysis module, disinfection effect evaluation module, control execution module, user interaction module and early warning response module; The data acquisition module uses a sensor network and video monitoring equipment to monitor and collect data on the nursing environment 24 hours a day to obtain nursing environment data; The data transmission module uses wireless technology and Ethernet to transmit the collected nursing environment data to the central server of the data processing and analysis module for processing; The data processing and analysis module uses a central server to store and process the nursing environment data, and uses a multi-agent-based communication and collaborative control method to perform data analysis and prediction to obtain nursing environment prediction results and control strategies; The disinfection effect evaluation module obtains the hospital environment nursing standards, which are used to evaluate the disinfection effects of wards, operating rooms and infection nursing areas in the nursing environment. According to the nursing environment data and hospital environment nursing standards obtained by the data acquisition module, a scoring system from 1 to 10 is used to evaluate the disinfection effects of different areas, where 1 indicates a very poor disinfection effect and 10 indicates an excellent disinfection effect, to obtain a disinfection effect score; The control execution module automatically adjusts the working state of each device in the nursing environment according to the obtained control strategy, and detects it through the disinfection effect evaluation module to obtain a disinfection effect score. According to the disinfection effect score, the control strategy is obtained through the data processing and analysis module to perform control adjustment; The user interaction module provides a graphical user interface to display monitored care environment data and system status information; The early warning response module sets a safety range for the nursing environment data, issues an alarm in a timely manner when the nursing environment data exceeds the safety range, and notifies medical staff through SMS, email, or APP reminders in the user interaction module.

2. The intelligent nursing environment monitoring and infection control system according to claim 1, characterized in that: The data processing and analysis module uses a multi-agent-based communication and collaborative control method to perform data analysis and prediction. The multi-agent-based communication and collaborative control method specifically includes the following steps: Step S1: Define the communication network model, take the devices in the sensor network as agents, and represent the communication relationship between agents as a binary communication graph ,in, is the set of intelligent agents, is the edge set, is the adjacency matrix; Step S2: Multi-agent communication: According to the communication network model, the agents process tasks and exchange information in order according to the topological sorting to obtain a solution; Step S3: Spatiotemporal graph conversion, converting the original binary communication graph Convert to a relaxation graph with continuous values , the formula used is as follows: ； ； In the formula, Represents a binary communication graph The adjacency matrix of and denote the spatial and temporal adjacency matrices respectively, and represents a differentiable graph mask, and denote the spatial and temporal relaxation graphs respectively, Represents the relaxation graph The adjacency matrix of Step S4: Optimize the relaxation graph spatiotemporal connectivity; Step S5: Control rule construction, using controllers, buffers and actuators to respond to data in an event-driven manner, perform automated control, and control rule design; Step S6: Define time delay ; Step S7: Set packet loss time limit, set limits on the frequency and duration of packet loss, define the time interval and , Indicates that the time delay is The total time length in the interval, Indicated in The total time length of the interval, where , Indicates the minimum delay, and Respectively and The upper bound of the time delay.

3. The intelligent nursing environment monitoring and infection control system according to claim 2 is characterized in that: In step S2, multi-agent communication specifically includes the following steps: Step S21: Initialization, setting the initial state and role of the agent, and determining the plug-in configuration of each agent; Step S22: query distribution, given a task q, distribute task q to all agents in sequence; Step S23: Topological sorting, sorting the dependencies between agents to obtain a directed acyclic spatial communication graph ; Step S24: message passing, each agent generates a response based on the task q, the currently set role, the state, and the information of its temporal neighbors and spatial neighbors. Temporal neighbors refer to the agents that have interacted with it in the previous round of dialogue, and spatial neighbors refer to the agents that have interacted with it in the same round of dialogue; Step S25: Iterate and update the summary, repeating steps S23 and S24 until M rounds of dialogue are completed and a final solution is generated.

4. The intelligent nursing environment monitoring and infection control system according to claim 2, characterized in that: In step S4, the relaxation graph is optimized The spatiotemporal connectivity of Step S41: Distribution approximation, using the policy gradient method to reflect the relaxation graph The underlying probability distribution of Step S42: low-rank sparsity, minimizing the nuclear norm of the graph mask, obtaining an optimized graph mask, and removing redundant information; Step S43: Perform a pruning operation on the optimized graph mask to remove redundant connections and obtain a binary mask matrix. The formula used is as follows: ； In the formula, is the characteristic function, is a binary mask matrix, represents the total number of elements in the adjacency matrix, represents the pruning ratio, Before selection elements, is a logical operator; Step S44: Update connectivity by applying the binary mask matrix to the adjacency matrix In the equation, we get the updated connectivity. The formula used is as follows: ； In the formula, and represents the spatial and temporal binary mask matrices, Represents a binary communication graph The subgraph of Represents the adjacency matrix of the subgraph after pruning optimization.

Images