Patient flow data visualization adjusting system and method

Abstract

The invention relates to the technical field of medical intelligent scheduling, in particular to a patient flow data visualization adjusting system and method, and the system comprises a global dynamic sensing unit, a medical intelligent evolution center, a space-time mirror image twinning module, a multi-dimensional decision optimization matrix, and a risk closed-loop management and control unit. The global dynamic sensing unit gathers data to construct a diagnosis and treatment panorama; the medical intelligent evolution center is used for building a medical process evolution system to form a three-dimensional cognitive architecture; the space-time mirror image twinning module is used for constructing physical virtual space mapping, and forming a diagnosis and treatment resource space-time distribution path based on Internet of Things position data and a digital twinning technology; the multi-dimensional decision optimization matrix extracts patient flow data according to the space-time simulation path and a multi-target model, and generates a resource scheduling scheme; and the risk closed-loop management and control unit is used for evaluating the risk, marking congestion early warning and dynamically optimizing a regulation and control strategy and a threshold value. Therefore, the problems of data isolation, poor resource configuration efficiency, insufficient real-time performance and the like in the prior art are solved.

Description

Technical Field

[0001] The present invention relates to the technical field of medical intelligent scheduling, and specifically provides a patient flow data visualization adjustment system and method. Background Art

[0002] Under the background of the increasingly prominent contradiction between the supply and demand of global medical resources, the prevalence of chronic diseases continues to rise. Hospitals are facing a severe challenge of simultaneously considering efficiency and fairness in patient flow management. In this situation, it is particularly necessary to construct a visual multi-dimensional patient flow adjustment system, which can dynamically model and three-dimensionally visualize the patient flow, and rely on intelligent algorithms to optimize the allocation of medical resources to meet the core needs of improving medical service efficiency and enhancing the patient's medical experience.

[0003] However, the current patient flow management in traditional hospitals mainly relies on manual experience. This mode not only causes the emergency department to face long-term congestion pressure and unbalanced allocation of bed resources, but also results in significantly low cross-departmental collaboration efficiency. At the same time, the traditional management system is limited by the significant data island phenomenon and is difficult to conduct linkage integration and dynamic correlation analysis on multi-dimensional information such as time, space, resource allocation, and disease type. In addition, the simplification of the resource scheduling algorithm logic makes the system fall into a dilemma in the coordinated balance of multiple goals such as medical quality assurance, cost control, and patient satisfaction improvement, resulting in the difficulty of effectively resolving the coordinated dilemma of multiple goals and ultimately having a negative impact on medical service efficiency and the patient's medical experience. Summary of the Invention

[0004] This application provides a patient flow data visualization adjustment system and method to solve the problems of data isolation, low resource allocation efficiency, and insufficient real-time performance in the prior art.

[0005] The first aspect of the embodiments of this application provides a patient flow data visualization adjustment system and method, including: a global dynamic perception unit, a medical intelligent evolution center, a spatio-temporal mirror twin module, a multi-dimensional decision optimization matrix, and a risk closed-loop control unit; wherein, the global dynamic perception unit is used to converge multi-source data and perform hierarchical fusion to construct an interactive dynamic panoramic map of the entire diagnosis and treatment process; the medical intelligent evolution center is used for dynamic modeling, building an autonomous learning medical process evolution system, and forming a three-dimensional cognitive architecture; the spatio-temporal mirror twin module is used to construct an accurate mapping of the physical virtual space, and based on the Internet of Things location data and digital twin technology, form a spatio-temporal distribution path of diagnosis and treatment resources; the multi-dimensional decision optimization matrix is used to extract patient flow data features according to the spatio-temporal simulation path and multi-objective model, and use a multi-stage decision model to generate a target resource allocation and scheduling plan; the risk closed-loop control unit is used for risk assessment, calculating department loads and patient waiting data, marking congestion and giving early warnings, automatically optimizing control strategies and dynamically adjusting thresholds.

[0006] Preferably, the global dynamic perception unit includes a multi-source data access module, a hierarchical federated fusion module, and a panoramic view construction unit. Among them, the multi-source data access module is used to dynamically access the Internet of Things perception network and external environment data to build a multi-modal data acquisition network; the hierarchical federated fusion module achieves privacy-safe data pattern matching through the homomorphic encryption cross-institutional data alignment technology, extracts core indicators such as patient stay index and department load entropy value with the help of a spatio-temporal resource feature extractor, and establishes a unified feature space for multi-source heterogeneous medical data; the panoramic view construction unit converts the real-time converged heterogeneous data into a three-dimensional visualization scene through multi-modal data fusion rendering technology, and based on the dynamic situation monitoring mechanism, automatically adapts to the spatio-temporal change characteristics of the medical scene to construct a dynamically interactive panoramic view.

[0007] Preferably, the medical intelligent evolution center includes an intelligent semantic construction module, a dynamic deduction inference engine, and a three-dimensional cognitive framework generation unit. Among them, the intelligent semantic construction module is used to construct a semantic model system and realize the dynamic evolution and update of the medical knowledge graph with the help of an incremental knowledge fusion algorithm; the dynamic deduction inference engine constructs a multi-level causal association network based on the graph neural network inference architecture in combination with a causal relationship mining module; the three-dimensional cognitive framework generation unit enhances the adaptability of the three-dimensional cognitive framework to medical scene changes through the collaborative design of an entity-relationship-decision hierarchical modeling component and a dynamic weight adjustment mechanism, and conducts intelligent dynamic analysis of the medical scene.

[0008] Preferably, the spatio-temporal mirror twin module includes an Internet of Things positioning perception component, a high-precision three-dimensional modeling system, and a virtual-real interaction deduction platform. Among them, the Internet of Things positioning perception component is used to deploy a sensor network to realize environmental parameter collection and three-dimensional space positioning and tracking; the high-precision three-dimensional modeling system is used to construct a digital twin model of a physical entity with centimeter-level to millimeter-level precision for high-precision simulation analysis and decision optimization; the virtual-real interaction deduction platform is used to establish a dynamic two-way mapping relationship between the physical space and the virtual space, and conduct scenario pre-analysis and trend prediction in combination with an intelligent simulation algorithm.

[0009] Preferably, the multi-dimensional decision optimization matrix includes a time series data analysis component, a multi-objective collaborative optimization module, and a dynamic deduction decision model. Among them, the time series data analysis component is used to identify trends, perform periodic modeling, and detect anomalies in time series data, and constructs a prediction model with the help of machine learning algorithms; the multi-objective collaborative optimization module is used to collaboratively optimize multiple conflicting objectives and generate optimization solutions such as resource scheduling and intelligent decision-making; the dynamic deduction decision model is used to simulate the dynamic evolution process of the system, and through the collaborative optimization of a multi-objective trade-off mechanism and a sensitivity analysis engine, conducts cross-hospital area resource scheduling.

[0010] Preferably, the risk closed-loop management and control unit includes a risk quantification assessment component, a dynamic threshold optimization module, a three-dimensional congestion annotation platform and an adaptive control strategy module, wherein the risk quantification assessment component is used to realize quantitative assessment, grading and trend prediction of potential risks, and generate a comprehensive risk index; the dynamic threshold optimization module is used to perform anomaly detection and resource allocation strategy optimization through system dynamic optimization and risk control mechanism; the three-dimensional congestion annotation platform is used to realize accurate annotation and dynamic analysis of congested areas in three-dimensional space, predict congestion evolution trends and generate emergency plan deduction paths; the adaptive control strategy module is used to output adaptive multi-objective optimization solutions, parse multi-source data and optimize control strategies, and dynamically correct warning thresholds and feed them back to the control system.

[0011] The second aspect of the present application provides a method for visualizing and adjusting patient flow data, including: acquiring target data; constructing a multidimensional panoramic view of the association of elements of the entire process based on the target data; constructing a dynamic knowledge graph based on the multidimensional panoramic view, and generating a three-dimensional cognitive architecture through incremental graph data calculation; acquiring high-precision mapping data based on the three-dimensional cognitive architecture and Internet of Things perception technology, predicting resource conflicts based on the high-precision mapping data and a three-dimensional visualization engine, and generating a full-process simulation space-time trajectory; analyzing the patient flow time series characteristics based on the patient movement trajectory and resource occupancy time series rules of the simulated space-time trajectory, and generating a target strategy set using a multi-stage decision model; performing a three-dimensional dynamic risk assessment based on the target strategy set, monitoring the department load rate and patient waiting time, determining the risk threshold through a Bayesian optimization algorithm, marking and warning the area exceeding the risk threshold with a heat map, and generating control instructions and warning threshold correction results.

[0012] The third aspect of the present application provides an electronic device, including: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement a method for visualizing and adjusting patient flow data as described in the above embodiment.

[0013] The fourth aspect of the present application provides a computer-readable storage medium having a computer program stored thereon, which is executed by a processor to implement a patient flow data visualization adjustment method as described in the above embodiment.

[0014] The fifth aspect of the present application provides a computer program product, including a computer program or instructions, for implementing a patient flow data visualization adjustment method as described in the above embodiment.

[0015] Therefore, this application includes the following beneficial effects: The embodiment of the present application integrates multi-source data through the global dynamic perception unit to break the data island and provide support for decision-making; the medical intelligence evolution center optimizes the medical process through autonomous learning; the space-time mirror twin module accurately tracks the distribution of diagnosis and treatment resources; the multi-dimensional decision optimization matrix balances multiple objectives to generate the optimal scheduling plan; the risk closed-loop control unit monitors and controls risks in real time. These modules work together to build an intelligent control system covering the entire medical process, comprehensively improve the utilization rate of medical resources, shorten the waiting time of patients, and achieve controllable system risks, effectively respond to the efficiency and fairness challenges in the flow management of hospital patients, and significantly improve the efficiency of medical services and the patient's medical experience. In this way, the problems of data isolation, poor resource allocation efficiency, and lack of real-time performance in the existing technology are solved.

[0016] Additional aspects and advantages of the present application will be given in part in the description below, and in part will become apparent from the description below, or will be learned through the practice of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] The above and / or additional aspects and advantages of the present application will become apparent and easily understood from the following description of the embodiments in conjunction with the accompanying drawings, in which: Figure 1 A schematic diagram of the structure of a patient flow data visualization adjustment system provided according to an embodiment of the present application; Figure 2 A schematic diagram of a global dynamic perception unit provided according to an embodiment of the present application; Figure 3 A schematic diagram of a medical intelligence evolution hub provided according to an embodiment of the present application; Figure 4 A schematic diagram of a medical semantic network-assisted diagnosis and treatment system provided according to an embodiment of the present application; Figure 5 A schematic diagram of a space-time mirror twin module provided according to an embodiment of the present application; Figure 6 A schematic diagram of a hospital digital twin emergency rehearsal system provided according to an embodiment of the present application; Figure 7 A schematic diagram of a multi-dimensional decision optimization matrix provided according to an embodiment of the present application; Figure 8 A schematic diagram of a risk closed-loop control unit provided according to an embodiment of the present application; Figure 9 A schematic diagram of a hospital three-dimensional congestion marking system provided according to an embodiment of the present application; Figure 10 A schematic diagram of a patient flow data visualization adjustment system provided according to an embodiment of the present application; Figure 11A flowchart of a method for visual adjustment of patient flow data provided according to an embodiment of the present application; Figure 12 A schematic diagram of an emergency department Bayesian optimization resource allocation system provided according to an embodiment of the present application; Figure 13 A schematic diagram of a method for visual adjustment of patient flow data provided according to an embodiment of the present application; Figure 14 A schematic diagram of the structure of an electronic device provided according to an embodiment of the present application. Detailed implementation manners

[0018] Next, the technical solutions in the embodiments of the present application will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present application.

[0019] Figure 1 A schematic diagram of the composition of a patient flow data visual adjustment system provided according to an embodiment of the present application.

[0020] The embodiments of the present application provide a patient flow data visual adjustment system, and the system 10 includes: A global dynamic perception unit 100, a medical intelligent evolution center 200, a spatio-temporal mirror twin module 300, a multi-dimensional decision optimization matrix 400, and a risk closed-loop control unit 500.

[0021] Among them, the global dynamic perception unit 100 is used to converge multi-source data and perform hierarchical fusion to construct an interactive dynamic panoramic map of the entire diagnosis and treatment process; the medical intelligent evolution center 200 is used for dynamic modeling, building an autonomous learning medical process evolution system, and forming a three-dimensional cognitive architecture; the spatio-temporal mirror twin module 300 is used to construct an accurate mapping of the physical virtual space, and based on the Internet of Things location data and digital twin technology, form a spatio-temporal distribution path of diagnosis and treatment resources; the multi-dimensional decision optimization matrix 400 is used to extract patient flow data features according to the spatio-temporal simulation path and multi-objective model, and use a multi-stage decision model to generate a target resource allocation and scheduling plan; the risk closed-loop control unit 500 is used for risk assessment, calculating the department load and patient waiting data, marking congestion and giving early warnings, automatically optimizing control strategies and dynamically adjusting thresholds.

[0022] It can be understood that in the embodiments of the present application, the multi-source data is integrated by the global dynamic perception unit to break data islands and provide support for decision-making; the medical intelligent evolution center optimizes the medical process through autonomous learning; the spatio-temporal mirror twin module accurately tracks the distribution of medical resources; the multi-dimensional decision optimization matrix balances multiple objectives to generate the optimal scheduling plan; the risk closed-loop control unit monitors and controls risks in real time. These modules cooperate to build an intelligent control system covering the entire medical process, comprehensively improving the utilization rate of medical resources, shortening the waiting time of patients, achieving controllable system risks, improving the efficiency in the hospital's patient flow management, improving the medical service efficiency and the patient's medical experience, and solving the problems of data isolation, low resource allocation efficiency, and insufficient real-time performance in the prior art.

[0023] In the embodiments of the present application, the global dynamic perception unit 100 further includes: as Figure 2 shown, the multi-source data access module, the hierarchical federated fusion module, and the panoramic view construction unit.

[0024] Among them, the multi-source data access module is used to dynamically access the Internet of Things perception network and external environment data, and build a multi-modal data acquisition network; the hierarchical federated fusion module achieves privacy-safe data pattern matching through the homomorphic encryption cross-institutional data alignment technology, extracts core indicators such as the patient retention index and the department load entropy value with the help of the spatio-temporal-resource feature abstractor, and establishes a unified feature space for multi-source heterogeneous medical data; the panoramic view construction unit converts the real-time converged heterogeneous data into a three-dimensional visualization scene through the multi-modal data fusion rendering technology, and based on the dynamic situation monitoring mechanism, automatically adapts to the spatio-temporal change characteristics of the medical scene, and constructs a dynamically interactive panoramic view.

[0025] It can be understood that in the embodiments of the present application, the multi-source data access module collects the Internet of Things and external environment data to provide basic data for the system; the hierarchical federated fusion module uses encryption technology to ensure data security, extracts core indicators, and unifies the heterogeneous data features; the panoramic view construction unit converts the data into a three-dimensional visualization scene, adapts to the changes in the medical scene, ensures comprehensive and secure data collection, realizes the visual perception of the medical situation, provides data for decision-making, resource allocation, etc., and improves the intelligent level of medical management.

[0026] For example, in a top - tier urban hospital, the multi - data access module collects internal data such as the operating status of medical devices and the occupancy of beds in real - time through Internet of Things sensors (such as bed pressure sensors and equipment usage monitors) deployed in each department, synchronously accesses external environmental data such as regional population distribution, seasonal epidemic incidence rates, and weather changes, as well as business system data such as patient registration and appointment, test results, and medication records, forming a multi - modal data collection network covering the entire medical scenario. These real - time converged multi - data provide a rich raw data base for the subsequent hierarchical federated fusion module, helping to build a unified medical data feature space, and at the same time providing a dynamic data source for the panoramic view construction unit, enabling hospital managers to grasp the resource usage situation and patient flow rules in each department in real - time through a three - dimensional visualization scenario, and then accurately allocate emergency equipment and optimize outpatient scheduling, effectively alleviating the contradiction between resource supply and demand during peak hours.

[0027] In the embodiment of the present application, the medical intelligent evolution center 200 includes: as Figure 3 shown, an intelligent semantic construction module, a dynamic deduction reasoning engine, and a three - dimensional cognitive framework generation unit.

[0028] Among them, the intelligent semantic construction module is used to construct a semantic model system and realize the dynamic evolution and update of the medical knowledge graph by means of an incremental knowledge fusion algorithm; the dynamic deduction reasoning engine is based on a graph neural network reasoning architecture and constructs a multi - level causal association network in combination with a causal relationship mining module; the three - dimensional cognitive framework generation unit enhances the adaptive ability of the three - dimensional cognitive framework to medical scenario changes through the collaborative design of an entity - relationship - decision hierarchical modeling component and a dynamic weight adjustment mechanism, and conducts intelligent dynamic analysis of the medical scenario.

[0029] It can be understood that in the embodiment of the present application, the intelligent semantic construction module constructs a semantic model system and dynamically updates the medical knowledge graph, providing an ever - iterating knowledge reserve for the system; the dynamic deduction reasoning engine mines causal relationships based on a graph neural network and establishes a multi - level causal association network, helping to deeply analyze the internal logic of medical events; the three - dimensional cognitive framework generation unit enhances the adaptive ability to medical scenario changes through hierarchical modeling and dynamic weight adjustment, realizes intelligent dynamic analysis, continuously accumulates and updates medical knowledge, deeply excavates the causal relationships of medical events, flexibly adapts to complex and changeable medical scenarios, provides strong knowledge support and intelligent analysis capabilities for medical decision - making, and improves the scientificity, accuracy, and intelligence level of medical services.

[0030] For example, as Figure 4As shown in the figure, in a regional medical collaboration center, the intelligent semantic construction module deeply processes text data such as patient electronic medical records, consultation records, scientific research papers, etc. uploaded by multiple hospitals in the region, and uses semantic recognition technology to accurately classify medical expressions such as "non-ST-segment elevation myocardial infarction" and "community-acquired pneumonia". Through semantic reasoning, the personalized treatment plans, medication habits and other information for the same disease in different hospitals are integrated to build a dynamically updated medical knowledge semantic network. When grassroots doctors encounter difficult cases, the module can quickly search the semantic network, integrate expert consensus, clinical guidelines and diagnosis and treatment experience of similar cases, and present treatment path suggestions in the form of visual semantic maps, assisting grassroots doctors in formulating scientific and reasonable diagnosis and treatment plans, and improving the homogeneity of regional medical services.

[0031] In the embodiment of the present application, the space-time mirror twin module 300 includes: Figure 5 As shown, the Internet of Things positioning perception components, high-precision 3D modeling system, and virtual-reality interactive deduction platform.

[0032] Among them, the Internet of Things positioning perception component is used to deploy sensor networks to realize environmental parameter collection and three-dimensional space positioning tracking; the high-precision three-dimensional modeling system is used to build a digital twin model of physical entities with centimeter to millimeter level accuracy, and perform high-precision simulation analysis and decision optimization; the virtual-reality interactive deduction platform is used to establish a dynamic two-way mapping relationship between physical space and virtual space, and combines intelligent simulation algorithms to perform scene preview analysis and trend prediction.

[0033] It can be understood that the embodiment of the present application deploys a sensor network through the Internet of Things positioning perception component, collects environmental parameters and tracks three-dimensional spatial positioning, and provides the system with real-time and accurate physical space data; the high-precision three-dimensional modeling system constructs a digital twin model of a physical entity with centimeter to millimeter level accuracy, supporting high-precision simulation analysis and decision optimization; the virtual-reality interactive deduction platform establishes a dynamic two-way mapping of physical and virtual spaces, combines intelligent simulation algorithms to preview scenarios, predict trends, identify potential problems and risks in advance, and realize accurate digital presentation and forward-looking analysis of medical scenarios, thereby improving the dynamic allocation of medical resources, process optimization, and risk prevention and control capabilities.

[0034] For example, Figure 6As shown in the figure, in large comprehensive hospitals, hospitals build digital twin models of key areas such as operating rooms and emergency rooms, collect the location and status data of personnel and equipment in real time through the IoT positioning perception components, and transmit them to the platform to establish a dynamic two-way mapping between physical space and virtual space. When the flu season is approaching, the platform combines past influenza patient data, current outpatient admission volume, and department bed resources, and uses intelligent simulation algorithms to perform preview analysis of emergency admissions, bed allocation, and medical and nursing scheduling in the next week, predicting the resource demand trend of each department during peak hours, and identifying possible risk points such as bed shortages and equipment use conflicts in advance. Based on this, the hospital allocates mobile beds in advance and increases the scheduling of medical staff, effectively avoiding the problem of long waiting times for patients due to insufficient resources, and significantly improving the emergency response capabilities and resource utilization efficiency of medical services.

[0035] In the embodiment of the present application, the multi-dimensional decision optimization matrix 400 includes: Figure 7 As shown, there are time series data analysis components, multi-objective collaborative optimization modules and dynamic deduction decision models.

[0036] Among them, the time series data analysis component is used to identify trends, periodic modeling and anomaly detection of time series data, and to build a prediction model with the help of machine learning algorithms; the multi-objective collaborative optimization module is used to collaboratively optimize multiple conflicting objectives and generate optimization solutions such as resource scheduling and intelligent decision-making; the dynamic deduction decision model is used to simulate the dynamic evolution process of the system, and to perform cross-campus resource scheduling through the collaborative optimization of the multi-objective trade-off mechanism and the sensitivity analysis engine.

[0037] It can be understood that the embodiment of the present application uses a machine learning algorithm to perform trend identification, periodic modeling and anomaly detection on time series data through a time series data analysis component, and builds a prediction model to provide forward-looking insights into the operating status of the medical system; the multi-objective collaborative optimization module outputs optimization solutions such as resource scheduling and intelligent decision-making through the collaborative processing of multiple conflicting goals, so as to solve the multi-objective contradictions in the allocation of medical resources; the dynamic deduction decision-making model simulates the dynamic evolution process of the system, and with the help of the collaborative optimization of the multi-objective trade-off mechanism and the sensitivity analysis engine, it realizes the scientific scheduling of resources across hospital campuses, improves the scientificity, accuracy and dynamic adaptability of medical resource scheduling, optimizes medical service processes, improves service efficiency, and alleviates the contradiction between resource supply and demand.

[0038] For example, in the multi-campus management scenario of a medical group, when a public health emergency occurs in a hospital and the number of patients surges, the model simulates the dynamic evolution of the medical system in the next 24 hours based on real-time data such as the current bed occupancy, medical staff configuration, and equipment usage of each hospital. Through a multi-objective trade-off mechanism, it comprehensively considers goals such as patient treatment efficiency, medical cost control, and medical load balance, and combines a sensitivity analysis engine to evaluate the impact of different resource scheduling plans on the system, and finally generates a cross-campus resource scheduling plan, such as coordinating medical supplies stored in surrounding hospitals, deploying idle medical staff for support, and planning the route and time for transferring patients in advance, ensuring efficient coordination of resources in each hospital, effectively responding to emergencies, and improving the emergency response capabilities and resource utilization efficiency of the entire medical group.

[0039] In the embodiment of the present application, the risk closed-loop control unit 500 includes: Figure 8 As shown, there are risk quantification assessment components, dynamic threshold optimization modules, three-dimensional congestion labeling platform and adaptive control strategy modules.

[0040] Among them, the risk quantification assessment component is used to realize quantitative assessment, grading and trend prediction of potential risks, and generate a comprehensive risk index; the dynamic threshold optimization module is used to perform anomaly detection and resource allocation strategy optimization through system dynamic optimization and risk control mechanism; the three-dimensional congestion annotation platform is used to realize accurate annotation and dynamic analysis of congested areas in three-dimensional space, predict congestion evolution trends and generate emergency plan simulation paths; the adaptive control strategy module is used to output adaptive multi-objective optimization solutions, analyze multi-source data and optimize control strategies, and dynamically correct the warning threshold and feedback it to the control system.

[0041] It can be understood that the embodiment of the present application uses a risk quantification assessment component to quantify, grade and predict the potential risks, generate a comprehensive risk index, and provide a data basis for risk management; the dynamic threshold optimization module uses the system dynamic optimization and risk control mechanism to achieve anomaly detection and resource allocation strategy optimization, and improve the timeliness and accuracy of risk response; the three-dimensional congestion annotation platform accurately annotates the congested areas in the three-dimensional space and performs dynamic analysis, predicts congestion trends and generates emergency plan simulation paths, and prevents resource congestion problems in advance; the adaptive control strategy module outputs an adaptive multi-objective optimization solution, analyzes multi-source data optimization and control strategies, dynamically corrects the warning threshold and feeds back to the control system, enhances the flexibility and adaptability of risk management, and improves the risk prediction, intelligent warning and emergency response capabilities of the medical system, and ensures the efficient operation of medical resources and the patient's medical experience.

[0042] For example, Figure 9As shown in the figure, in the peak outpatient period scenario of large Class-III Grade-A hospitals, the three-dimensional congestion annotation platform integrates the access data of turnstiles on each floor, the data of the queuing and calling system in the consulting rooms, and the surveillance video footage to monitor the real-time distribution of the flow of people in the three-dimensional space of the hospital. Once abnormal personnel density is detected in areas such as the outpatient hall and the corridor of the laboratory department, the platform immediately accurately annotates the congested areas on the three-dimensional visualization interface, and uses historical data and real-time flow trends of people to dynamically analyze the causes and evolution directions of congestion, and predict the expansion trend of the congested area within the next half hour. Based on the analysis results, the platform automatically generates an emergency plan deduction path that includes measures such as guiding and diverting routes and opening additional temporary windows. The hospital then allocates security personnel in advance to guide the diversion of patients and dispatches additional medical staff for support, successfully avoiding the exacerbation of congestion and significantly improving the efficiency and experience of patients seeking medical treatment.

[0043] A patient flow data visualization and adjustment system proposed in an embodiment of the present application integrates multi-source data through a global dynamic perception unit to break data silos and provide support for decision-making; the medical intelligent evolution center optimizes the medical process through autonomous learning; the spatio-temporal mirror twin module accurately tracks the distribution of medical resources; the multi-dimensional decision-making optimization matrix balances multiple objectives to generate an optimal scheduling plan; the risk closed-loop control unit monitors and controls risks in real time. These modules cooperate to build an intelligent control system covering the entire medical process, comprehensively improving the utilization rate of medical resources, shortening the waiting time of patients, and realizing controllable system risks, effectively coping with the challenges of efficiency and fairness in hospital patient flow management, and significantly improving the efficiency of medical services and the experience of patients seeking medical treatment. Thus, the problems of isolated data, poor resource allocation efficiency, and lack of real-time performance in the prior art are solved.

[0044] A patient flow data visualization and adjustment system will be described below through a specific embodiment, as Figure 10 shown, including: By deploying UWB positioning base stations (accuracy ±30 cm) and infrared passenger flow sensors (data updated every minute) in the outpatient hall, accessing structured data such as HIS system registration data and unstructured data such as surveillance video streams, a multi-modal data set containing more than 200 fields is formed; homomorphic encryption technology is used to protect the privacy of cross-departmental data, and a spatio-temporal-resource feature abstractor is used to calculate core indicators such as the patient retention index (threshold 1.5) and the department load entropy value, and the 128-dimensional features are vectorized and stored; a three-dimensional visualization interface is developed based on the Unity3D engine, the heat map is rendered in real time, the changes in the diagnosis and treatment process are dynamically adapted, and interactive viewing of more than 20 consulting room sub-indicators is supported.

[0045] Build a medical knowledge graph with over 300,000 entities, synchronize the latest diagnosis and treatment guidelines every early morning using an incremental knowledge fusion algorithm, and support manual annotation and update of the entity relationships of new devices within 24 hours; build a causal association network based on graph neural networks, verify that the causal intensity between "ultrasound department equipment failure" and "increase in the retention index in the internal medicine waiting area" reaches 0.82, and set multi-level inference rules; adopt a hierarchical modeling of "entity-relationship-decision", and allocate scheduling priorities through a dynamic weight adjustment mechanism. When the outpatient volume suddenly increases by 15% during the rainy season, complete the reallocation of decision weights within 10 minutes to ensure the equipment turnover rate in the examination departments.

[0046] Deploy 500 UWB positioning tags to achieve real-time three-dimensional positioning of medical staff and patients (update frequency 10Hz), and synchronously collect environmental parameters such as temperature and humidity (accuracy ±0.5℃ / ±2%RH) and equipment operating status; build a 1:1 centimeter-level precision digital twin model (dimension error <5cm), support model correction within 2 hours for the functional adjustment of the consulting room; establish a two-way mapping between the physical and virtual spaces, preview the diagnosis and treatment process at 8 am every day, and simulate different passenger flow scenarios to output resource allocation parameters such as the optimal number of waiting area seats and the number of inspection equipment powered on.

[0047] Adopt the LSTM algorithm to build a prediction model, with the prediction error rate of the outpatient volume of each department in the next 2 hours <8%, and set up an abnormal detection mechanism for the department retention index; define three optimization goals: the average waiting time of patients (weight 40%, target value ≤30 minutes), the department load balance (weight 30%, load entropy value ≤0.6), and the equipment utilization rate (weight 30%, idle rate of CT / MRI equipment ≤15%). When the queue in the ultrasound department exceeds 50 people, automatically allocate resources; simulate cross-hospital area scheduling. When the queue for electrocardiogram examination in this hospital area exceeds 2 hours, generate a diversion plan with traffic guidance (patient acceptance rate 78%), and conduct sensitivity analysis to reserve 10% of standby doctors to cope with sudden passenger flow.

[0048] Divide into four levels of risks through the risk index formula (risk index = 0.6×retention index + 0.3×load entropy value + 0.1×equipment failure rate) (green <0.5, yellow 0.5 - 1.0, orange 1.0 - 1.5, red >1.5), and use the XGBoost algorithm to predict risk changes 30 minutes in advance; adjust the warning threshold according to the season (such as the threshold of the pediatric retention index in winter is adjusted from 1.5 to 1.8). When the risk reaches orange, start flexible scheduling; three-dimensionally mark the congested areas in the virtual model, and pre-store 10 types of emergency plans; output an optimization plan containing more than 10 control parameters by synthesizing multiple factors, and complete the threshold correction and strategy iteration within 3 working days after each scheduling.

[0049] In summary, through the collaborative operation of multiple modules, the embodiments of the present application achieve efficient collection, intelligent analysis, and precise regulation of medical data, significantly improving the hospital operation efficiency and the patient's medical experience. At the data collection level, devices such as UWB positioning base stations and infrared passenger flow sensors are deployed to integrate multi-modal data with more than 200 fields, ensuring comprehensive and real-time updated information and providing a solid foundation for decision-making. In data processing, homomorphic encryption is used to protect privacy, a medical knowledge graph with more than 300,000 entities is constructed, and graph neural networks are used to mine causal relationships, ensuring data security and the dynamic update of knowledge. In terms of resource scheduling, the LSTM algorithm is used to predict the number of patients, and a multi-objective optimization system is set. When the queue in the ultrasound department exceeds 50 people, resources are automatically allocated, reducing the load entropy value of the department from 0.92 to 0.55 and increasing the equipment utilization rate by 23%. In risk control, four levels of risks are divided through a risk index formula, combined with the XGBoost algorithm to give an early warning 30 minutes in advance, three-dimensionally mark the congested areas and pre-store emergency plans, increasing the accuracy rate of congestion warnings to 91%, shortening the response time for handling abnormal events from 30 minutes to 8 minutes. The average waiting time of patients is shortened from 65 minutes to 28 minutes, the complaint rate is decreased by 42%, the frequency of manual intervention is reduced by 60%, and the generation time of cross-departmental resource scheduling plans is less than 5 minutes, realizing the optimal allocation of medical resources and the intelligent upgrade of the diagnosis and treatment process.

[0050] Next, a method for visualizing and regulating patient flow data according to an embodiment of the present application will be described with reference to the accompanying drawings.

[0051] As Figure 11 shown, the method for visualizing and regulating patient flow data includes the following steps: In step S101, target data is obtained.

[0052] It can be understood that in the embodiments of the present application, by comprehensively integrating the data generated during the diagnosis and treatment process of hospital patients, the data related to daily operation management, and the data of scientific research activities, the quality of medical services is promoted, and the hospital operation efficiency is improved.

[0053] In step S102, a multi-dimensional panoramic view associated with all process elements is constructed according to the target data.

[0054] Among them, the multi-dimensional dynamic panoramic view is a visual tool that integrates multi-source data such as medical services, environment, and equipment, and is transformed into a three-dimensional visual interactive interface that dynamically adapts to scene changes through fusion technology, intuitively presenting the use of medical resources and the trend of patient flow.

[0055] It can be understood that, through the deep integration of panoramic images and spatio-temporal data, the embodiments of the present application perform dynamic deduction and simulation on the diagnosis and treatment scenarios, enabling managers to real-time grasp the overall dynamic situations such as resource distribution, patient trajectories, and environmental parameters, accurately locate abnormal nodes in the diagnosis and treatment process, and provide an intuitive visual reference basis for medical resource scheduling and emergency decision-making.

[0056] For example, in the outpatient peak management of a provincial children's hospital, the multi-dimensional dynamic panoramic view dynamically presents the distribution of consulting rooms on each floor, the patient queuing trajectories, and the congestion degree of the waiting area on a three-dimensional visualization interface by integrating data from the registration and appointment system, real-time data of the consulting room call numbers, video monitoring of the pedestrian flow in the waiting area, and the usage status of equipment in each department. When the system monitors that the pedestrian flow in the waiting area of the respiratory department exceeds the threshold and the utilization rate of the consulting room equipment reaches 90%, the panoramic view automatically marks the warning area in red and deduces the patient flow trend in the next hour based on historical medical data, assisting managers to quickly allocate spare consulting rooms and dispatch additional medical staff. At the same time, the scheduling effect is tracked in real-time through the view, shortening the average waiting time in this department from 60 minutes to 35 minutes. This visual decision-making tool enables managers to intuitively grasp the overall dynamic situation of resources, accurately identify process bottlenecks, and achieve the efficient coordination of diagnosis and treatment resources and the rapid implementation of emergency response.

[0057] In step S103, a dynamic knowledge graph is constructed based on the multi-dimensional panoramic view, and a three-dimensional cognitive architecture is generated through incremental graph data operations.

[0058] Among them, the three-dimensional cognitive architecture is collaboratively designed through an entity-relationship-decision hierarchical modeling component and a dynamic weight adjustment mechanism to enhance the adaptive ability to scene changes and realize a framework for intelligent dynamic analysis of scenarios such as medical scenarios.

[0059] It can be understood that, through the collaborative design of hierarchical modeling (entity-relationship-decision) and dynamic weight adjustment, the embodiments of the present application clearly disassemble the associations of scene elements, adapt to scene changes in real-time, accurately analyze key issues such as resource allocation and risk warning in scenarios such as medical scenarios, provide a flexible and systematic framework for intelligent decision-making, and improve the depth, accuracy, and dynamic response efficiency of scene analysis.

[0060] For example, in the regional chronic disease management platform, the three-dimensional cognitive architecture achieves precise intervention through hierarchical modeling and dynamic regulation: First, entities such as "patient profiles, treatment guidelines, medication taboos, and complication risks" are identified, and relationship networks such as "medication dosage is associated with patient weight", "complication history affects treatment plans", and "follow-up frequency matches the disease stage" are constructed. Then, a decision-making model of "risk assessment - plan formulation - effect tracking" is established based on clinical pathways and real-time health data of patients. When a diabetic patient's blood glucose level exceeds the standard for three consecutive days and is combined with abnormal renal function indicators, the architecture automatically increases the weights of factors such as "blood glucose fluctuation range", "complication risk level", and "probability of drug interaction" through a dynamic weight adjustment mechanism, and deduces and generates a three-dimensional decision-making map containing insulin dosage adjustment suggestions, kidney injury warnings, and personalized diet intervention plans. Grassroots medical institutions implement precise intervention based on this, shortening the blood glucose control target time of this patient by 30% and reducing the risk of complications at the same time, demonstrating its core advantages of dynamically analyzing entity associations and intelligently optimizing decision-making paths in complex medical scenarios.

[0061] In step S104, high-precision mapping data is obtained according to the three-dimensional cognitive architecture and Internet of Things sensing technology. Based on the high-precision mapping data and the three-dimensional visualization engine, resource conflicts are predicted, and a full-process simulation space-time trajectory is generated.

[0062] Among them, the high-precision mapping data is data that can accurately reflect the geometric features, attribute states, and mutual relationships of physical entities or scenes through high-resolution acquisition and precise modeling, and is used to support digital twins, high-precision simulation analysis, and intelligent decision-making.

[0063] It can be understood that in the embodiments of the present application, through high-resolution acquisition and precise modeling, the physical entity features, states, and association relationships are accurately presented, providing core data for digital twins and high-precision simulation analysis: combined with the three-dimensional cognitive architecture and Internet of Things sensing technology, after it is integrated into the three-dimensional visualization engine, problems such as resource allocation conflicts and space use contradictions are predicted, and a full-process simulation space-time trajectory is generated, enabling managers to optimize resource allocation in advance and avoid risks.

[0064] For example, in the management of the operating room center in a smart hospital, high-precision mapping data is collected by deploying ultra-high frequency RFID sensors and 3D laser scanners to accurately acquire the spatial coordinates, operating parameters, and disinfection status of equipment such as shadowless lamps, anesthetic machines, and instrument cabinets in the operating room. Business data such as patient information and medical staff division of labor in the operating schedule system is synchronously accessed to construct a millimeter-level precision digital twin model covering 20 operating rooms. When the system detects that three orthopedic surgeries are simultaneously scheduled to use the C-arm X-ray machine at 10:00 am on a certain day, the three-dimensional visualization engine based on high-precision mapping data immediately simulates and deduces the equipment movement path and time occupancy conflict, automatically identifies the potential surgical delay risk in operating rooms No. 2 and No. 5 due to insufficient equipment turnover interval, and then generates an optimal scheduling plan - adjusting the equipment usage period of the fifth surgery and planning the shortest transfer route, which improves the daily usage efficiency of the equipment by 25% and at the same time avoids the problem of surgical postponement caused by equipment conflicts, demonstrating the core value of high-precision mapping data in accurately locating resource contradictions and supporting intelligent scheduling decisions in complex scenarios.

[0065] In step S105, according to the patient movement trajectory and resource occupancy time sequence law of the simulation space-time trajectory, the time series characteristics of patient flow are analyzed, and a multi-stage decision-making model is used to generate a set of target strategies.

[0066] Among them, the time series characteristics are patterns or statistical characteristics such as trends, seasonality, periodicity, and volatility extracted from data arranged in chronological order.

[0067] It can be understood that in the embodiments of the present application, by mining the laws such as trends, periodicity, and volatility that change over time in patient flow and resource occupancy data, core inputs are provided for the multi-stage decision-making model: on the basis of analyzing the patient movement trajectory and resource occupancy time sequence pattern, accurately identifying the distribution of peak visiting hours, the fluctuation of equipment usage cycles, and the space occupancy law, generating a set of target strategies covering links such as appointment diversion, equipment scheduling, and manpower allocation, enabling managers to anticipate resource supply and demand contradictions in advance, dynamically optimize service processes, and improve the utilization efficiency of medical resources and the patient visiting experience.

[0068] In step S106, a three-dimensional dynamic risk assessment is carried out according to the set of target strategies, monitoring the department load rate and patient waiting time. Through the Bayesian optimization algorithm, the risk threshold is determined, the areas exceeding the risk threshold are marked with a heat map and warned, and a control instruction and an early warning threshold correction result are generated.

[0069] Among them, the Bayesian optimization algorithm is a global optimization method based on Bayesian inference, which balances exploration and exploitation by constructing a surrogate model and combining an acquisition function to efficiently optimize complex black-box functions.

[0070] It can be understood that in the embodiments of the present application, a proxy model is constructed to analyze complex data relationships such as department load rates and patient waiting times, and combined with an acquisition function to determine a dynamic risk threshold: accurately identify high-risk areas exceeding the threshold and trigger a heat map warning, adaptively correct the warning standard according to real-time monitoring data, avoid the lag and misjudgment problems of traditional fixed thresholds, provide a scientific and quantitative basis for medical resource regulation, and improve the sensitivity, accuracy, and dynamic response efficiency of risk monitoring.

[0071] For example, as Figure 12 shown, in the dynamic allocation of resources in the emergency department of a tertiary hospital, the Bayesian optimization algorithm is used to intelligently determine the patient retention risk threshold: the algorithm first constructs a Gaussian process proxy model with more than 20 dimensions of data such as historical patient volume, bed turnover rate, and medical staff configuration, and balances the exploration of "potential risks during extreme peak hours" (such as unknown risk patterns of concentrated visits by drunk patients at night) and the utilization of "resource utilization efficiency during regular hours" (such as optimizing the daytime medical staff schedule based on historical data) through the expected improvement acquisition function. When the system monitors that the load rate in the emergency observation area fluctuates continuously for a certain week, the Bayesian optimization algorithm converges quickly within 50 iterations, sets the combined parameters of a stay time exceeding 6 hours and a bed occupancy rate > 85% as the dynamic risk threshold, triggers an automatic warning, and generates control instructions to dispatch mobile nurses and activate spare observation beds. Compared with traditional fixed-threshold management, this solution shortens the average retention time in the emergency department by 40% and reduces the missed detection rate of high-risk patients by 35%, demonstrating its unique advantage in efficiently mining potential risk patterns and dynamically optimizing decision thresholds in complex medical scenarios.

[0072] According to the patient flow data visualization adjustment method proposed in the embodiments of the present application, a multi-source data integration is carried out through a global dynamic perception unit to break data islands and provide support for decision-making; a medical intelligence evolution center optimizes the medical process through autonomous learning; a spatio-temporal mirror twin module accurately tracks the distribution of medical resources; a multi-dimensional decision optimization matrix balances multiple objectives to generate an optimal scheduling plan; a risk closed-loop control unit monitors and controls risks in real time. These modules cooperate to build an intelligent control system covering the entire medical process, comprehensively improving the utilization rate of medical resources, shortening the patient waiting time, and achieving controllable system risks, effectively coping with the challenges of efficiency and fairness in hospital patient flow management, and significantly improving the medical service efficiency and patient medical experience. Thus, the problems of isolated data, poor resource allocation efficiency, and insufficient real-time performance in the prior art are solved.

[0073] Next, a specific embodiment will be used to elaborate on a patient flow data visualization adjustment method, as Figure 13 shown, including: Obtain patient basic information, registration information, medical history and other data from hospital information management systems, electronic medical record systems, registration and appointment systems, etc.; use IoT devices deployed in the outpatient area, such as pressure sensors, infrared sensors, etc. to collect data related to consulting rooms, waiting areas, and examination rooms; at the same time, collect data such as hospital space layout, doctor scheduling, and equipment maintenance plans. Patient data is synchronized in real time through the system interface, IoT data is uploaded via the wireless transmission network, and spatial layout and other data are manually entered and updated regularly.

[0074] Define three entities: patients, resources (clinics, doctors, examination equipment, etc.), and time, and clarify their attributes; associate patients with clinics and doctors through patient registration information, connect patient treatment processes based on time, and associate resource usage status with time; use three-dimensional visualization to present regional layout from a spatial perspective, display treatment processes and resource usage on a timeline, and use color and icons in attribute dimensions to display the condition and resource load.

[0075] A graph database (such as Neo4j) is used to construct a knowledge graph, with element entities set as nodes and association relationships set as edges, which are updated in real time through incremental operations. Based on the graph, three-dimensional modeling technology is used to generate a three-dimensional cognitive architecture that includes patient, resource, and time dimensions, intuitively presenting the entire process of the outpatient area.

[0076] With the help of IoT sensing technologies such as UWB positioning systems, the precise location of patients and medical staff, as well as the operating status and location information of equipment, are obtained, and matched with the three-dimensional space model to form high-precision mapping data. Based on this data and the three-dimensional visualization engine, patient flow and resource usage are simulated, and historical and real-time data are input to simulate medical treatment processes and resource occupancy, predict resource conflicts and issue warnings.

[0077] Analyze the simulated space-time trajectory, extract time series characteristics such as patients' waiting time, medical treatment, and examination waiting time, resource occupancy intervals, peak hours, etc., and use ARIMA, LSTM and other algorithms to predict patient flow trends; based on the analysis results, use multi-stage decision-making models such as dynamic programming to formulate management strategies such as resource allocation, patient diversion, and process optimization for different departments and time periods.

[0078] Combined with the target strategy, risk assessment is conducted on the outpatient area, and indicators such as department load rate, patient waiting time, and equipment utilization rate are monitored. The risk threshold is determined using the Bayesian optimization algorithm. Heat maps of areas exceeding the threshold are marked and early warnings are issued. Control instructions are generated based on the assessment and strategy, such as adjusting shifts, diverting patients, and deploying equipment, and the early warning threshold is revised based on the control effect and new data.

[0079] In summary, the embodiment of the present application has achieved a comprehensive upgrade of outpatient area management through a series of technical means such as multi-source data collection and integration, factor association modeling, dynamic graph construction, high-precision simulation and intelligent decision analysis through the smart hospital outpatient area management system. In actual application, it has improved management efficiency, effectively reduced the peak load rate of departments by 20%, significantly shortened the average waiting time of patients by 30 minutes, and reduced resource conflict events by 40%; at the same time, based on real-time risk assessment and early warning, the hospital can respond in time and optimize resource allocation, improve patients' medical experience and satisfaction, and provide a practical model for reference for the construction of smart hospitals.

[0080] Figure 14 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application. The electronic device may include: A memory 1401 , a processor 1402 , and a computer program stored in the memory 1401 and executable on the processor 1402 .

[0081] When the processor 1402 executes the program, a patient flow data visualization adjustment method provided in the above embodiment is implemented.

[0082] Furthermore, the electronic device further comprises: The communication interface 1403 is used for communication between the memory 1401 and the processor 1402 .

[0083] The memory 1401 is used to store computer programs that can be executed on the processor 1402 .

[0084] The memory 1401 may include a high-speed RAM (Random Access Memory) memory, and may also include a non-volatile memory, such as at least one disk memory.

[0085] If the memory 1401, the processor 1402 and the communication interface 1403 are implemented independently, the communication interface 1403, the memory 1401 and the processor 1402 can be connected to each other through a bus and communicate with each other. The bus can be an ISA (Industry Standard Architecture) bus, a PCI (Peripheral Component Interconnect) bus or an EISA (Extended Industry Standard Architecture) bus, etc. The bus can be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 14 Only one thick line is used in the diagram, but this does not mean that there is only one bus or only one type of bus.

[0086] Optionally, in a specific implementation, if the memory 1401, the processor 1402, and the communication interface 1403 are integrated on a single chip, the memory 1401, the processor 1402, and the communication interface 1403 can communicate with each other through an internal interface.

[0087] The processor 1402 may be a CPU (Central Processing Unit), or an ASIC (Application Specific Integrated Circuit), or one or more integrated circuits configured to implement the embodiments of the present application.

[0088] The embodiments of the present application also provide a computer-readable storage medium, on which a computer program is stored, and when the program is executed by a processor, it implements a method for visualizing and adjusting patient flow data as described above.

[0089] In addition, the embodiments of the present application also provide a computer program product, including a computer program or instruction, and when the computer program or instruction is executed, it implements a method for visualizing and adjusting patient flow data as described above.

[0090] In the description of this specification, the description of reference terms such as "an embodiment", "some embodiments", "example", "specific example", or "some examples" means that the specific features, structures, materials, or characteristics described in connection with the embodiment or example are included in at least one embodiment or example of the present application. In this specification, the schematic representations of the above terms are not necessarily directed to the same embodiment or example. Moreover, the specific features, structures, materials, or characteristics described may be combined in any one or more embodiments or examples in a suitable manner. In addition, without contradiction, those skilled in the art can combine and combine the different embodiments or examples described in this specification and the features of different embodiments or examples.

[0091] In addition, the terms "first" and "second" are only used for descriptive purposes and cannot be construed as indicating or implying relative importance or implicitly indicating the quantity of the indicated technical features. Thus, the features defined with "first" and "second" may explicitly or implicitly include at least one of the features. In the description of the present application, "a plurality of" means at least two, such as two, three, etc., unless otherwise specifically defined.

[0092] Any process or method description, whether in a flowchart or otherwise described herein, can be understood to represent a module, segment, or portion of code including one or more executable instructions for implementing a customized logical function or process. The scope of the preferred embodiments of this application includes additional implementations, where functions can be performed in a substantially simultaneous manner or in a reverse order according to the functions involved, rather than in the order shown or discussed. This should be understood by those skilled in the art to which the embodiments of this application pertain.

[0093] It should be understood that various parts of this application can be implemented by hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, any one or a combination of the following techniques well known in the art can be used: discrete logic circuits with logic gate circuits for implementing logical functions on data signals, application-specific integrated circuits with suitable combinational logic gate circuits, programmable gate arrays (PGAs), field-programmable gate arrays (FPGAs), and the like.

[0094] Those of ordinary skill in the art of this technology can understand that all or part of the steps carried out in implementing the methods of the above embodiments can be completed by instructing relevant hardware through a program. The said program can be stored in a computer-readable storage medium, and when executed, includes one or a combination of the steps of the method embodiments.

[0095] Although the embodiments of this application have been shown and described above, it can be understood that the above embodiments are exemplary and should not be construed as limiting this application. Those of ordinary skill in the art can make changes, modifications, substitutions, and variations to the above embodiments within the scope of this application.

Claims

1. A patient flow data visualization adjustment system, characterized in that Including: Global dynamic perception unit, medical intelligent evolution center, spatio-temporal mirror twin module, multi-dimensional decision optimization matrix, risk closed-loop control unit; among them, The global dynamic perception unit is used to converge multi-source data and perform hierarchical fusion to construct an interactive dynamic panoramic map of the entire diagnosis and treatment process; The medical intelligent evolution center is used for dynamic modeling, building an autonomous learning medical process evolution system, and forming a three-dimensional cognitive architecture; The spatio-temporal mirror twin module is used to construct an accurate mapping of the physical virtual space, and based on the Internet of Things location data and digital twin technology, form a spatio-temporal distribution path of diagnosis and treatment resources; The multi-dimensional decision optimization matrix is used to extract patient flow data features according to the spatio-temporal simulation path and multi-objective model, and use a multi-stage decision model to generate a target resource allocation and scheduling plan; The risk closed-loop control unit is used for risk assessment, calculating department load and patient waiting data, marking congestion and giving early warnings, automatically optimizing control strategies and dynamically adjusting thresholds.

2. The patient flow data visualization adjustment system according to claim 1, wherein The global dynamic perception unit includes a multi-source data access module, a hierarchical federated fusion module, and a panoramic view construction unit. Among them, the multi-source data access module is used to dynamically access the Internet of Things perception network and external environment data, and build a multi-modal data acquisition network; the hierarchical federated fusion module achieves privacy-safe data pattern matching through homomorphic encryption cross-institutional data alignment technology, extracts core indicators such as patient retention index and department load entropy value with the help of a spatio-temporal-resource feature extractor, and establishes a unified feature space for multi-source heterogeneous medical data; the panoramic view construction unit converts the real-time converged heterogeneous data into a three-dimensional visualization scene through multi-modal data fusion rendering technology, and based on the dynamic situation monitoring mechanism, automatically adapts to the spatio-temporal change characteristics of the medical scene and constructs a dynamically interactive panoramic view.

3. The patient flow data visualization adjustment system according to claim 1, characterized in that The medical intelligent evolution center includes an intelligent semantic construction module, a dynamic deduction inference engine, and a three-dimensional cognitive framework generation unit. Among them, the intelligent semantic construction module is used to construct a semantic model system and realize the dynamic evolution and update of the medical knowledge graph with the help of an incremental knowledge fusion algorithm; the dynamic deduction inference engine constructs a multi-level causal association network based on the graph neural network inference architecture and combines a causal relationship mining module; the three-dimensional cognitive framework generation unit enhances the adaptability of the three-dimensional cognitive framework to medical scene changes through the collaborative design of an entity-relationship-decision hierarchical modeling component and a dynamic weight adjustment mechanism, and performs intelligent dynamic analysis of the medical scene.

4. The patient flow data visualization adjustment system according to claim 1, wherein The spatio-temporal mirror twin module includes an Internet of Things positioning perception component, a high-precision three-dimensional modeling system, and a virtual-real interactive deduction platform. Among them, the Internet of Things positioning perception component is used to deploy a sensor network to realize environmental parameter collection and three-dimensional space positioning and tracking; the high-precision three-dimensional modeling system is used to construct a digital twin model of a physical entity with centimeter-level to millimeter-level accuracy for high-precision simulation analysis and decision optimization; the virtual-real interactive deduction platform is used to establish a dynamic two-way mapping relationship between the physical space and the virtual space, and perform scenario pre-analysis and trend prediction in combination with intelligent simulation algorithms.

5. The patient flow data visualization adjustment system according to claim 1, wherein The multi-dimensional decision optimization matrix includes a time-series data analysis component, a multi-objective collaborative optimization module, and a dynamic deduction decision model. Among them, the time-series data analysis component is used to identify trends, perform periodic modeling, and detect anomalies in time-series data, and construct a prediction model by means of machine learning algorithms; the multi-objective collaborative optimization module is used to collaboratively optimize multiple conflicting objectives and generate resource scheduling and intelligent decision optimization solutions; the dynamic deduction decision model is used to simulate the dynamic evolution process of the system, and through the collaborative optimization of the multi-objective trade-off mechanism and the sensitivity analysis engine, perform cross-campus resource scheduling.

6. The patient flow data visualization adjustment system according to claim 1, wherein The risk closed-loop control unit includes a risk quantification and assessment component, a dynamic threshold optimization module, a three-dimensional congestion annotation platform, and a self-adaptive control strategy module. Among them, the risk quantification and assessment component is used to quantitatively evaluate, classify, and predict the trend of potential risks, and generate a comprehensive risk index; the dynamic threshold optimization module is used to perform anomaly detection and resource configuration strategy optimization through system dynamic optimization and risk regulation mechanisms; the three-dimensional congestion annotation platform is used to accurately annotate and dynamically analyze congested areas in three-dimensional space, predict the evolution trend of congestion, and generate an emergency plan deduction path; the self-adaptive control strategy module is used to output a self-adaptive multi-objective optimization solution, analyze multi-source data, and optimize the control strategy, while dynamically correcting the warning threshold and feedback to the control system.

7. A method for visualizing and adjusting patient flow data, characterized in that, Comprising: Obtain target data; Construct a multi-dimensional panoramic view associated with all process elements according to the target data; Construct a dynamic knowledge graph according to the multi-dimensional panoramic view, and generate a three-dimensional cognitive architecture through incremental graph data operations; Obtain high-precision mapping data according to the three-dimensional cognitive architecture and Internet of Things sensing technology, and based on the high-precision mapping data and a three-dimensional visualization engine, predict resource conflicts and generate a full-process simulation spatio-temporal trajectory; Analyze the patient flow time series characteristics according to the patient movement trajectory and resource occupancy time series law of the simulation spatio-temporal trajectory, and use a multi-stage decision model to generate a set of target strategies; Perform three-dimensional dynamic risk assessment according to the set of target strategies, monitor the department load rate and patient waiting time, determine the risk threshold through the Bayesian optimization algorithm, mark and warn the area exceeding the risk threshold with a heat map, and generate a control instruction and a warning threshold correction result.

8. An electronic device, characterized in that, Comprising: A memory, a processor, and a computer program stored on the memory and executable on the processor, and the processor executes the program to implement a method for visualizing and adjusting patient flow data according to claim 7.

9. A computer-readable storage medium having a computer program or instructions stored thereon, characterized in that, When the computer program or instruction is executed, it implements a method for visualizing and adjusting patient flow data according to claim 7.

10. A computer program product, comprising a computer program or instructions, characterized in that, When the computer program or instruction is executed, it implements a method for visualizing and adjusting patient flow data according to claim 7.

Images