Building space layout optimization method and system based on medical building body

Abstract

The invention relates to the field of building space layout, and discloses a building space layout optimization method and system based on a medical building ontology, and the method comprises the steps: constructing a medical building field ontology library; acquiring related data of the medical building; performing semantic mapping, cleaning and fusion on the medical building related data according to the medical building domain ontology library to form a medical building digital twin knowledge graph; based on semantic association information carried by the medical building digital twin knowledge graph, constructing a spatial relationship quantitative model used for evaluating spatial layout performance; generating a plurality of spatial layout schemes on the basis of the spatial relationship quantification model under the condition that a preset constraint condition is met; and performing performance evaluation on each spatial layout scheme based on the spatial relationship quantitative model, and obtaining a spatial layout scheme of which the evaluation result meets a preset optimization target from the plurality of spatial layout schemes. According to the technical scheme, the problem that building design and medical operation data are disjointed in the prior art is solved.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of architectural space layout, and in particular to a method and system for optimizing architectural space layout based on medical building ontology. BACKGROUND

[0002] In the design process of medical buildings, the rationality of space layout affects the efficiency of medical processes, the experience of patients seeking medical treatment, and the level of hospital infection control.

[0003] An ideal space organization should make high-frequency collaborative functional units, such as outpatient departments and pharmacies, emergency departments and imaging departments, adjacent in space and meet the standard constraints such as clean and dirty separation.

[0004] However, in current design practice, architectural design is disconnected from medical operation data, and design in the design stage mainly relies on the experience of architects and static standards, and it is difficult to integrate real business process data generated by hospital information systems (HIS), electronic medical records (EMR), etc., resulting in a lack of forward-looking verification of the actual running state after the design is completed, and there is a risk that the design is disconnected from the process.

[0005] How to solve this technical problem is a technical problem that needs to be overcome by those skilled in the art. SUMMARY

[0006] The embodiments of the present application provide a method for optimizing architectural space layout based on medical building ontology to at least partially solve the above technical problems.

[0007] In order to achieve the above-mentioned purpose, according to the first aspect of the present application, a method for optimizing architectural space layout based on medical building ontology is provided, comprising: constructing a medical building domain ontology library; the medical building domain ontology library is used to define functional space units, medical business activities, personnel roles, equipment facilities in the medical building and the semantic relationships between them; obtaining medical building related data; the medical building related data includes building and environment data and medical and operation data; performing semantic mapping, cleaning and fusion on the medical building related data according to the medical building domain ontology library to form a medical building digital twin knowledge graph; based on the semantic association information carried by the medical building digital twin knowledge graph, constructing a space relationship quantitative model for evaluating the performance of space layout; based on the space relationship quantitative model, generating a plurality of space layout schemes under the condition of meeting the preset constraint conditions; based on the space relationship quantitative model, performing performance evaluation on each space layout scheme, and obtaining a space layout scheme whose evaluation result meets the preset optimization target from the plurality of space layout schemes.

[0008] Optionally, based on the semantic association information carried by the medical building digital twin knowledge graph, a spatial relationship quantitative model for evaluating the performance of the spatial layout is constructed, comprising: extracting operation data representing the collaborative relationship between different functional units from the medical building digital twin knowledge graph; the operation data includes the flow frequency of personnel between different functional units and the connection frequency of medical business processes between functional units; calculating the functional association strength between any two functional units based on the operation data and generating a functional association strength matrix; obtaining a digital building model formed by the building and environmental data for simulation analysis; associating the medical building digital twin knowledge graph with the digital building model; deploying a multi-agent simulation environment in the digital building model; simulating the movement process of patients, medical staff and materials in the building space according to a predetermined spatial layout scheme; based on the simulation results, obtaining at least one spatiotemporal performance parameter in the line efficiency index, the average moving distance or the space congestion point distribution; the functional association strength matrix and the spatiotemporal performance parameter are jointly used as input to construct a spatial relationship quantitative model for comprehensive evaluation of the advantages and disadvantages of the spatial layout.

[0009] Optionally, calculating the functional association strength between any two functional units based on the operation data and generating a functional association strength matrix, comprising: for each pair of functional units, respectively obtaining the personnel flow frequency value and the medical business process connection frequency value corresponding thereto; normalizing the personnel flow frequency value and the medical business process connection frequency value to obtain normalized personnel flow strength values and normalized business connection strength values; configuring a first weight coefficient for the normalized personnel flow strength value and a second weight coefficient for the normalized business connection strength value; the sum of the first weight coefficient and the second weight coefficient is 1; multiplying the first weight coefficient by the normalized personnel flow strength value and adding the product of the second weight coefficient and the normalized business connection strength value to obtain a comprehensive functional association strength value corresponding to the functional unit pair; traverse all functional unit pairs, taking each functional unit as row index and column index; fill the corresponding comprehensive functional association strength value into the corresponding position of the matrix to generate a functional association strength matrix.

[0010] Optionally, a medical building domain ontology library is constructed, and the medical building domain ontology library is used to define functional space units, medical business activities, personnel roles, equipment facilities, and semantic relationships therebetween in the medical building. Medical building related data is acquired, and the medical building related data includes building and environment data and medical and operation data. The medical building related data is subjected to semantic mapping, cleaning, and fusion according to the medical building domain ontology library, to form a medical building digital twin knowledge graph. Based on semantic association information carried by the medical building digital twin knowledge graph, a spatial relationship quantification model for evaluating spatial layout performance is constructed. Based on the spatial relationship quantification model, a plurality of spatial layout schemes are generated under the condition of satisfying a preset constraint condition. Based on the spatial relationship quantification model, performance of each spatial layout scheme is evaluated, and a spatial layout scheme whose evaluation result satisfies a preset optimization target is obtained from the plurality of spatial layout schemes.

[0011] Optionally, according to a preset spatial layout scheme, a moving process of patients, medical staff, and materials in a building space is simulated, including: Based on the preset spatial layout scheme, specific arrangement positions of each functional unit in the building space are determined. The arrangement positions are mapped to the digital building model to generate a simulation scene corresponding to the spatial layout scheme. Intelligent agents respectively representing patients, medical staff, and materials are configured in the simulation scene, and behavior rules thereof are set. The intelligent agents are driven to run in the simulation scene to simulate their moving process under the spatial layout scheme.

[0012] Optionally, the medical building related data is subjected to semantic mapping, cleaning, and fusion according to the medical building domain ontology library to form a medical building digital twin knowledge graph, including: Geometric information and spatial attributes are extracted from the building and environment data; the geometric information includes room number, area, and coordinates; and the spatial attributes include functional partition and floor distribution. Business process information and personnel role information are extracted from the medical and operation data; the business process information includes diagnosis and treatment process and equipment use record; and the personnel role information includes medical staff responsibility and patient treatment path. The geometric information, spatial attributes, business process information, and personnel role information are subjected to semantic mapping to generate a unified data structure according to definitions in the medical building domain ontology library. data cleaning is performed on the unified data structure to remove duplicate data, error data or inconsistent data; Based on the pre-defined semantic relationship in the medical building domain ontology library, the cleaned semantic entities are associated; semantic links are established between functional space units, medical business activities, personnel roles and equipment facilities to form knowledge graph nodes and edges; The knowledge graph nodes and edges are stored in a graph database to construct a medical building digital twin knowledge graph.

[0013] Optionally, the functional correlation strength matrix and the space-time performance parameter are jointly used as inputs to construct a space relationship quantification model for comprehensive evaluation of the advantages and disadvantages of the space layout, including: According to the pre-set space layout scheme, the geometric positions of each functional unit in the digital building model are determined; the spatial distance between any two functional units is calculated to generate a spatial distance matrix; The functional correlation strength matrix and the spatial distance matrix are multiplied element by element to obtain a product matrix of correlation strength and spatial distance; Sum all the non-diagonal elements in the product matrix of correlation strength and spatial distance to obtain a layout correlation cost value; The layout correlation cost value and the space-time performance parameter are fused to form a comprehensive evaluation index; wherein the fusion method includes using the space-time performance parameter as a hard constraint condition, and only when the pre-set threshold is met, the layout scheme is accepted; or, the layout correlation cost value and each space-time performance parameter are respectively configured with a weight coefficient, and a weighted comprehensive score is calculated; The comprehensive evaluation index is used as the output of the space relationship quantification model.

[0014] Optionally, based on the space relationship quantification model, a plurality of space layout schemes are generated under the condition of meeting the pre-set constraint condition, including: S801, extracting a set of functional units to be laid out from the medical building digital twin knowledge graph and obtaining the functional correlation strength between any two functional units; S802, determining the space occupancy parameter of each functional unit according to the area requirement of each functional unit and the total amount of available space; S803, generating an initial space layout scheme by reducing the total sum of the product of functional correlation strength and spatial distance combined with the pre-set constraint condition; S804, inputting the initial space layout scheme into the space relationship quantification model to obtain the corresponding layout correlation cost value and space-time performance parameter; fusing the layout correlation cost value and the space-time performance parameter to form a comprehensive evaluation index; S805, if the comprehensive evaluation index does not reach the preset diversity or convergence condition, performing a perturbation operation on the current layout scheme, generating a new candidate layout scheme and returning to S804 for reevaluation; S804 and S805 are repeated until a plurality of spatial layout schemes satisfying the preset constraint conditions and different comprehensive evaluation indexes are obtained.

[0015] Optionally, performance of each spatial layout scheme is evaluated based on the spatial relationship quantification model, and a spatial layout scheme with an evaluation result satisfying a preset optimization target is obtained from the plurality of spatial layout schemes. The spatial relationship quantification model is called to evaluate the plurality of spatial layout schemes respectively, and a comprehensive evaluation index corresponding to each scheme is obtained. The comprehensive evaluation indexes of the plurality of spatial layout schemes are sorted or screened according to the preset optimization target; wherein the preset optimization target includes at least one of the following: minimizing the layout correlation cost value, maximizing the flow efficiency, and ensuring that all spatiotemporal performance parameters satisfy the corresponding preset threshold value. A spatial layout scheme with the optimal comprehensive evaluation index and satisfying all preset constraint conditions is selected as the final optimization result.

[0016] According to a second aspect of the present application, a medical building spatial layout optimization system based on ontology and knowledge graph is provided, comprising: A first processing module is configured to construct a medical building domain ontology library; the medical building domain ontology library is used to define functional space units, medical business activities, personnel roles, equipment facilities in the medical building, and semantic relationships therebetween. A second processing module is configured to obtain medical building related data; the medical building related data includes building and environment data and medical and operation data. A third processing module is configured to perform semantic mapping, cleaning and fusion on the medical building related data according to the medical building domain ontology library, and form a medical building digital twin knowledge graph. A fourth processing module is configured to construct a spatial relationship quantification model for evaluating spatial layout performance based on semantic association information carried by the medical building digital twin knowledge graph. A fifth processing module is configured to generate a plurality of spatial layout schemes under the condition of satisfying preset constraint conditions based on the spatial relationship quantification model. A sixth processing module is configured to evaluate the performance of each spatial layout scheme based on the spatial relationship quantification model, and obtain a spatial layout scheme with an evaluation result satisfying a preset optimization target from the plurality of spatial layout schemes.

[0017] In summary, the embodiment of the present application solves the problem of disconnection between architectural design and medical operation data in the prior art through the above technical solution.

[0018] Other features and advantages of the present application will be described in detail in the following detailed description section. BRIEF DESCRIPTION OF DRAWINGS

[0019] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings needed in the embodiment description will be briefly introduced. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor on the basis of these drawings.

[0020] Figure 1 is a step flow chart of a building space layout optimization method based on medical building ontology provided in an exemplary embodiment of the present application; Figure 2 is a system schematic diagram of a medical building space layout optimization system based on ontology and knowledge graph provided in an exemplary embodiment of the present application; Explanation of reference numerals: 01, first processing module; 202, second processing module; 203, third processing module; 204, fourth processing module; 205, fifth processing module; 206, sixth processing module. DETAILED DESCRIPTION

[0021] The technical solutions in the embodiments of the present application will be described clearly and completely in the following with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only some of the embodiments of the present application, not all. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor are within the protection scope of the present application.

[0022] The present application provides a building space layout optimization method based on medical building ontology, please refer to Figure 1 The building space layout optimization method based on medical building ontology provided in the embodiment of the present application includes the following steps: Step 101, constructing a medical building field ontology library; the medical building field ontology library is used to define the functional space units, medical business activities, personnel roles, equipment facilities and the semantic relationships between them in the medical building.

[0023] Specifically, the medical building domain ontology library is a machine-readable knowledge model built with a standard ontology language such as OWL, and its core role is to provide a unified semantic definition and relationship framework for all key concepts in the medical building domain. The functional space unit refers to a physical space with specific medical functions, such as an outpatient clinic, operating room, pharmacy, and inpatient ward. Medical business activities refer to medical behaviors that occur in these spaces, such as registration, consultation, examination, surgery, and medication. Personnel roles include participants of different identities such as patients, doctors, nurses, and pharmacists. Equipment and facilities include CT machines, MRIs, beds, and call systems.

[0024] Step 102, acquiring medical building related data; the medical building related data includes building and environment data and medical operation data.

[0025] Specifically, the building and environment data comes from the digital system of the whole life cycle of the building, including but not limited to: building information model data, computer-aided design drawings, building automation systems, and real-time environmental parameters such as real-time passenger flow, temperature and humidity collected by Internet of Things sensors; the building information model data contains the geometry, material and attribute information of the building components. Medical and operation data comes from the information system of the hospital, including registration, charging, scheduling records generated by the hospital information system; diagnosis and treatment process data in the electronic medical record; test application and result data of the laboratory information system.

[0026] Step 103, according to the medical building domain ontology library, the medical building related data is semantically mapped, cleaned and fused to form a medical building digital twin knowledge graph.

[0027] Specifically, in order to realize high-quality knowledge fusion, the application adopts the strategy of data hierarchical governance. The core of this strategy is to regard the multi-source heterogeneous raw data collected from BIM, HIS, IoT and other systems as raw materials to be governed, and to perform systematic semantic mapping, data cleaning and deep fusion on the raw data according to the unified semantic standard provided by the aforementioned ontology library in the medical building field. Through this hierarchical governance process, the original chaotic and isolated data stream is transformed into structured and machine-understandable knowledge. The heterogeneous data obtained in step 102 is semantically mapped using the ontology library constructed in step 101, for example, a room object in the BIM model is associated with the functional space unit concept of the imaging department-CT room in the ontology library through its spatial semantic tag such as the CT scan room; the business log of patient A in the HIS system completing the examination in the CT room is mapped to the semantic link between patient A, CT examination activity and CT room in the ontology library. The mapped data is cleaned to remove duplicate, incorrect or inconsistent records, and all cleaned and standardized semantic entities and their relationships are stored in a graph database to construct a medical building digital twin knowledge graph.

[0028] Step 104, based on the semantic association information carried by the medical building digital twin knowledge graph, a spatial relationship quantization model for evaluating the performance of the space layout is constructed.

[0029] Specifically, this step is used to convert abstract semantic knowledge into computable evaluation indicators; specifically, operational data representing the collaborative closeness between functional units are extracted from the knowledge graph, for example, the frequency of patients flowing from the internal medicine outpatient department to the outpatient pharmacy, or the frequency of process connection of transferring patients from the emergency department to the surgery department in the historical data; based on these data, the functional association strength between any two functional units is calculated and a functional association strength matrix is generated; the knowledge graph is associated with the digital building model constructed by BIM and other data, so that each functional unit can correspond to a specific three-dimensional spatial position; on this basis, a multi-agent simulation environment is deployed, and by simulating the movement of a large number of virtual patients, medical staff and material agents under the preset layout, spatiotemporal performance parameters such as average route efficiency, total moving distance and spatial congestion point distribution are obtained; the functional association strength matrix and the spatiotemporal performance parameters are jointly used as inputs to construct a spatial relationship quantization model, which can output a numerical indicator to measure the pros and cons of any space layout scheme.

[0030] Step 105, based on the spatial relationship quantization model, a plurality of space layout schemes are generated under the condition of meeting the preset constraint conditions.

[0031] Specifically, the system takes a set of functional units in the knowledge graph as a starting point, generates an initial layout scheme using a spatial arrangement algorithm such as a genetic algorithm under the premise of meeting preset constraint conditions such as clean- dirty flow line separation and minimum functional unit area.

[0032] Step 106, performance evaluation of each spatial layout scheme based on the spatial relationship quantitative model, obtaining a spatial layout scheme from the multiple spatial layout schemes whose evaluation result meets a preset optimization target.

[0033] Specifically, the present application solves the problems in the prior art that building design is disconnected from medical operation data, data value is difficult to mine due to semantic islands, and spatial relationship evaluation is subjective and lacks quantitative basis, thereby improving the forward-looking nature of design.

[0034] In some embodiments, based on the semantic association information carried by the medical building digital twin knowledge graph, a spatial relationship quantitative model for evaluating the performance of spatial layout is constructed, including: extracting operation data representing the cooperative relationship between different functional units from the medical building digital twin knowledge graph; the operation data includes the flow frequency of personnel between different functional units and the connection frequency of medical business processes between functional units; calculating the functional association strength between any two functional units based on the operation data and generating a functional association strength matrix; obtaining a digital building model formed by the building and environmental data for simulation analysis; associating the medical building digital twin knowledge graph with the digital building model; deploying a multi-agent simulation environment in the digital building model; simulating the movement process of patients, medical staff and materials in the building space according to a preset spatial layout scheme; based on the simulation result, obtaining at least one spatiotemporal performance parameter from the line efficiency index, average moving distance or space congestion point distribution; taking the functional association strength matrix and the spatiotemporal performance parameter as inputs, constructing a spatial relationship quantitative model for comprehensive evaluation of the pros and cons of spatial layout.

[0035] Specifically, operation data representing the cooperative relationship between different functional units is extracted from the medical building digital twin knowledge graph; specifically, these operation data are behavior evidence obtained by the knowledge graph through long-term fusion of historical logs of HIS, EMR and other systems, which mainly include: flow frequency of personnel between different functional units and connection frequency of medical business processes between functional units.

[0036] Based on the operational data, the system calculates the functional association strength between any two functional units and generates a functional association strength matrix. Specifically, the system normalizes the extracted personnel flow frequency and business process connection frequency to eliminate differences in units, and performs weighted fusion according to preset weight coefficients (e.g., personnel flow accounts for 70%, process connection accounts for 30%) to obtain a functional association strength value between 0 and 1. The higher the value, the closer the business collaboration between the two functional units. By traversing all pairs of functional units, an N×N functional association strength matrix is ​​generated; N is the total number of functional units, and the elements in the matrix are... This represents the strength of the association between the i-th and j-th functional units.

[0037] A digital building model for simulation analysis is obtained from the building and environmental data; and the digital twin knowledge graph of the medical building is associated with the digital building model. Specifically, the digital building model is a three-dimensional geometric and topological model constructed with BIM data as the core, describing the building's spatial structure, walls, doors, windows, and room boundaries. The key to associating the knowledge graph with this model lies in establishing a semantic-geometric mapping. By matching the semantic definitions of functional units in the knowledge graph with the spatial semantic tags of spatial objects in the digital building model, each functional unit is bound to its specific geometric position in three-dimensional space.

[0038] A multi-agent simulation environment is deployed within the digital building model. Based on a preset spatial layout scheme, it simulates the movement of patients, medical staff, and supplies within the building space. This multi-agent simulation environment is a simulation technology based on individual behavior rules. In this environment, the system configures three types of agents: patient agents, medical staff agents, and supply agents. Patient agents follow a standard or variant path: registration → waiting → consultation → examination → medication pickup. Medical staff agents operate in fixed areas according to their shifts or conduct cross-area consultations. Supply agents, such as those handling medicines and specimens, move between the pharmacy, laboratory, and patient rooms. When a preset spatial layout scheme is input, the simulation engine drives hundreds or thousands of agents to run concurrently within the digital building model, recreating the scene after the building is put into use.

[0039] Based on the simulation results, at least one spatiotemporal performance parameter is obtained from the following: mobility efficiency index, average travel distance, or spatial congestion point distribution. Specifically, after the simulation, the system statistically analyzes the agent's trajectory and outputs a series of spatiotemporal performance parameters. The mobility efficiency index is defined as the ratio of effective service time to total dwell time. The average travel distance refers to the average path length required for various types of personnel to complete a typical task. The spatial congestion point distribution is identified by a heat map showing areas where the flow of people exceeds a threshold per unit time.

[0040] Using the functional correlation strength matrix and the spatiotemporal performance parameters as inputs, a spatial relationship quantification model is constructed to comprehensively evaluate the merits and demerits of spatial layout.

[0041] In some embodiments, calculating the functional association strength between any two functional units based on the operational data and generating a functional association strength matrix includes: For each pair of functional units, obtain the corresponding personnel flow frequency value and medical business process connection frequency value; The personnel mobility frequency and medical business process connection frequency values ​​are normalized to obtain normalized personnel mobility intensity and normalized business connection intensity values. A first weighting coefficient is configured for the normalized personnel mobility intensity value, and a second weighting coefficient is configured for the normalized business connection intensity value; the sum of the first weighting coefficient and the second weighting coefficient is 1. Multiply the first weighting coefficient by the normalized personnel mobility intensity value and add the product of the second weighting coefficient and the normalized business connection intensity value to obtain the comprehensive functional association intensity value of the corresponding functional unit pair. Traverse all functional unit pairs, using each functional unit as the row index and column index; fill the corresponding comprehensive functional association strength value into the corresponding position in the matrix to generate the functional association strength matrix.

[0042] Specifically, for each pair of functional units, the corresponding personnel flow frequency and medical business process connection frequency values ​​are obtained. In particular, the system queries historical operation records from the digital twin knowledge graph of the medical building to count the personnel flow frequency values ​​from functional unit A to functional unit B within a preset time period; and counts the medical business process connection frequency values ​​of medical activities initiated by functional unit A triggering a response from functional unit B, such as the number of instances where a CT examination request form is issued in the outpatient department and then a CT scan is performed in the radiology department.

[0043] The frequency values ​​of personnel movement and the frequency values ​​of medical business process connections are normalized to obtain normalized personnel movement intensity values ​​and normalized business connection intensity values. Specifically, since the dimensions and orders of magnitude of the original data may differ greatly—for example, the frequency of personnel movement may be in the thousands, while a specific business connection may only be in the hundreds—directly adding them together will lead to one type of data dominating the result. Methods such as min-max normalization or Z-score normalization can be used to map the two types of original values ​​to the interval [0, 1], thereby obtaining dimensionless normalized personnel movement intensity values ​​and normalized business connection intensity values.

[0044] A first weighting coefficient is assigned to the normalized personnel flow intensity value, and a second weighting coefficient is assigned to the normalized business connection intensity value; the sum of the first weighting coefficient and the second weighting coefficient is 1; the first weighting coefficient and the second weighting coefficient are used to reflect the relative importance that hospital managers or designers attach to the two types of collaborative evidence; for example, in the design of outpatient areas centered on patient experience, personnel flow can be given a higher weight; while in surgical centers that emphasize process standardization, the connection logic of business processes may be more important.

[0045] Multiply the first weighting coefficient by the normalized personnel mobility intensity value and add the product of the second weighting coefficient and the normalized business connection intensity value to obtain the comprehensive functional association intensity value of the corresponding functional unit pair.

[0046] Iterate through all functional unit pairs, using each functional unit as its row and column index; fill the corresponding comprehensive functional association strength value into the appropriate position in the matrix to generate a functional association strength matrix; assuming there are N functional units to be laid out in the current design task, construct an N x N square matrix. The rows and columns of the matrix are arranged in the same order for these N functional units, and the element in the i-th row and j-th column is the value calculated above. .

[0047] In some embodiments, associating the digital twin knowledge graph of the medical building with the digital building model includes: Extract spatial objects with independent functional attributes from the digital building model; obtain the unique geometric identifier and spatial semantic label for each spatial object; Semantic definitions of each functional unit are extracted from the digital twin knowledge graph of the medical building; the semantic definitions include at least one of function type, department, and service role. Based on the matching of the spatial semantic tags with the semantic definitions of the functional units, a one-to-one correspondence between spatial objects and functional units is established. The one-to-one correspondence is stored in the medical building digital twin knowledge graph as a link attribute, so that any functional unit can be associated with a specific spatial geometric location in the digital building model.

[0048] Specifically, spatial objects with independent functional attributes are extracted from the digital building model; a unique geometric identifier and spatial semantic label are obtained for each spatial object. Specifically, the digital building model contains a large number of building components, and this step focuses on identifying the smallest spatial unit with complete medical functions, i.e., a spatial object, such as a specific consultation room, an operating room, or a pharmacy counter area. The system extracts two key attributes for each spatial object: one is a unique geometric identifier, such as an Element ID or IFC GUID in BIM, used to locate the object in the 3D model; the other is a spatial semantic label, which is a description of the functional use assigned to the space during the architectural design phase, such as attributes like room type being an internal medicine consultation room, floor F2, and area being 15 square meters.

[0049] Semantic definitions of each functional unit are extracted from the digital twin knowledge graph of the medical building; the semantic definition includes at least one of function type, department, and service role; specifically, the functional unit is a node in the knowledge graph representing medical business capabilities, and its semantic definition is pre-defined by the medical building domain ontology library.

[0050] Based on the matching of the spatial semantic tags with the semantic definitions of the functional units, a one-to-one correspondence between spatial objects and functional units is established. Specifically, the system performs a semantic alignment operation: comparing the spatial semantic tags of a spatial object in the digital building model with the semantic definitions of all functional units in the knowledge graph; when the two reach a consensus on key fields, it is determined that the spatial object carries the business activities of the functional unit, thereby establishing a one-to-one correspondence between the two; this process can be achieved with the help of a rule engine or lightweight natural language processing technology, and manual verification by designers is supported when necessary.

[0051] The one-to-one correspondence is stored in the medical building digital twin knowledge graph as a link attribute, so that any functional unit can be associated with a specific spatial geometric location in the digital building model; specifically, the system adds a link attribute to each functional unit node in the knowledge graph, the value of which is the unique geometric identifier of the corresponding spatial object.

[0052] In some embodiments, the movement of patients, medical staff, and supplies within a building space is simulated according to a preset spatial layout scheme, including: Based on the preset spatial layout scheme, the specific location of each functional unit in the building space is determined; The arrangement locations are mapped to the digital building model to generate a simulation scene corresponding to the spatial layout scheme; In the simulation scenario, intelligent agents representing patients, medical staff, and supplies are configured, and their behavioral rules are set. The agent is driven to run in the simulation scenario, simulating its movement process under the spatial layout scheme.

[0053] Specifically, based on the preset spatial layout scheme, the specific location of each functional unit in the building space is determined; specifically, the preset spatial layout scheme refers to a spatial organization form to be evaluated in the current optimization iteration, which clearly defines the geometric occupancy area of ​​each functional unit in the building plane or three-dimensional space, for example, by defining its position and boundary through coordinate range, room number list or polygon outline.

[0054] The layout locations are mapped onto the digital building model to generate a simulation scene corresponding to the spatial layout scheme. Specifically, the layout location of each functional unit in the layout scheme is matched and bound to the corresponding physical space area in the digital building model. If it is a completely new design scheme with no existing BIM room, a virtual room geometry that meets the area requirements can be generated within the available space of the digital building model. After mapping, a complete simulation scene consistent with the current layout scheme is constructed, which includes the spatial location, connectivity paths, and environmental attributes of all functional units. Intelligent agents representing patients, medical staff, and supplies are configured in the simulation scene, and their behavioral rules are set.

[0055] The intelligent agent is driven to run in the simulation scenario, simulating its movement process under the spatial layout scheme; specifically, the simulation engine advances in discrete time steps, generates intelligent agents in batches according to preset initial conditions, such as daily visit volume and staff schedule, and autonomously navigates, avoids obstacles, queues and interacts in the geometric space of the simulation scenario according to its behavior rules.

[0056] In some embodiments, semantic mapping, cleaning, and fusion of medical building-related data are performed based on the medical building domain ontology to form a medical building digital twin knowledge graph, including: Geometric information and spatial attributes are extracted from the building and environmental data; the geometric information includes room number, area, and coordinates; the spatial attributes include functional zoning and floor distribution. Business process information and personnel role information are extracted from the medical and operational data; the business process information includes the diagnosis and treatment process and equipment usage records; the personnel role information includes the responsibilities of medical staff and the patient's medical treatment path. Based on the definitions in the ontology library of the medical building domain, the geometric information, spatial attributes, business process information and personnel role information are semantically mapped to generate a unified data structure. The unified data structure is cleaned to remove duplicate, erroneous, or inconsistent data. Based on the predefined semantic relationships in the ontology library of the medical building domain, the cleaned semantic entities are associated; semantic links are established between functional space units, medical business activities, personnel roles and equipment facilities to form knowledge graph nodes and edges; The knowledge graph nodes and edges are stored in a graph database to construct a digital twin knowledge graph for medical buildings.

[0057] In some embodiments, the geometric position of each functional unit in the digital building model is determined according to the preset spatial layout scheme; the spatial distance between any two functional units is calculated to generate a spatial distance matrix; specifically, in the current preset spatial layout scheme to be evaluated, each functional unit has been assigned to a specific area within the building space. Using established semantic-geometric links, the geometric center point or representative coordinates of each functional unit in the digital building model are obtained. Based on the actual passageway within the building, the shortest path algorithm is used to calculate the spatial distance between any two functional units, i.e., the actual path length required for a person to move between them. The distance values ​​between all N functional units are filled into an N×N square matrix to obtain the spatial distance matrix, where the elements... This represents the travel distance from functional unit i to j.

[0058] The functional association strength matrix and the spatial distance matrix are multiplied element by element to obtain the product matrix of association strength and spatial distance; specifically, the functional association strength matrix is ​​multiplied by the corresponding elements of the spatial distance matrix to generate a new N×N matrix, which is the product matrix of association strength and spatial distance.

[0059] The layout association cost value is obtained by summing all off-diagonal elements in the product matrix of the association strength and spatial distance. Since the diagonal elements (i=j) represent the same functional unit itself, their distance is 0, so they have no practical meaning. The layout association cost value is obtained by summing all off-diagonal elements in the matrix (i.e., all items i≠j). This value can be understood as the total efficiency loss of the entire layout scheme due to the mismatch between business collaboration needs and spatial proximity. The smaller the value, the closer the highly associated departments are, and the more the layout conforms to the business logic.

[0060] The layout-related cost value and the spatiotemporal performance parameters are integrated to form a comprehensive evaluation index. The integration method includes using the spatiotemporal performance parameters as hard constraints, accepting the layout scheme only when a preset threshold is met; or assigning weight coefficients to the layout-related cost value and each spatiotemporal performance parameter, and calculating a weighted comprehensive score. Constraint mode: Key spatiotemporal parameters, such as the maximum number of congestion points, are set as hard constraints. Only when the simulation results meet a preset threshold is the layout scheme considered feasible, and its layout-related cost value is included in the comparison. Weighted mode: Weight coefficients are assigned to the layout-related cost value and each spatiotemporal performance parameter, and a weighted comprehensive score is calculated through linear weighting. The comprehensive evaluation index is used as the output of the spatial relationship quantification model.

[0061] In some embodiments, based on the spatial relationship quantification model, multiple spatial layout schemes are generated under preset constraints, including: S801. Extract the set of functional units to be laid out from the digital twin knowledge graph of the medical building and obtain the functional association strength between any two functional units. S802. Determine the space occupancy parameters for each functional unit based on the area requirements of each functional unit and the total available space in the building. S803. With the goal of reducing the sum of the product of functional correlation strength and spatial distance, and combined with preset constraints, an initial spatial layout scheme is generated. S804. Input the initial spatial layout scheme into the spatial relationship quantification model to obtain the corresponding layout association cost value and spatiotemporal performance parameters; integrate the layout association cost value and the spatiotemporal performance parameters to form a comprehensive evaluation index; S805. If the comprehensive evaluation index does not meet the preset diversity or convergence conditions, the current layout scheme is perturbed to generate a new candidate layout scheme and the process is returned to S804 for re-evaluation. Repeat steps S804 and S805 until multiple spatial layout schemes that meet the preset constraints and have different comprehensive evaluation indicators are obtained.

[0062] In some embodiments, performance evaluation of each spatial layout scheme is performed based on the spatial relationship quantification model, and a spatial layout scheme whose evaluation results meet the preset optimization target is obtained from the plurality of spatial layout schemes, including: The spatial relationship quantification model is invoked to evaluate the multiple spatial layout schemes respectively, and a comprehensive evaluation index corresponding to each scheme is obtained; The comprehensive evaluation indicators of the multiple spatial layout schemes are sorted or filtered according to the preset optimization objectives; wherein, the preset optimization objectives include at least one of the following: minimizing the layout-related cost value, maximizing circulation efficiency, and ensuring that all spatiotemporal performance parameters meet the corresponding preset thresholds; The spatial layout scheme with the best comprehensive evaluation index and that meets all preset constraints is selected as the final optimization result.

[0063] Reference Figure 2 The second embodiment of the present invention provides a medical building spatial layout optimization system based on ontology and knowledge graph, including: The first processing module 201 is used to: construct an ontology library for the medical building domain; the ontology library for the medical building domain is used to define functional space units, medical business activities, personnel roles, equipment and facilities in medical buildings, and the semantic relationships between them. The second processing module 202 is used to: acquire medical building-related data; the medical building-related data includes building and environmental data as well as medical and operational data; The third processing module 203 is used to: perform semantic mapping, cleaning and fusion of the medical building-related data based on the medical building ontology library to form a digital twin knowledge graph of medical buildings; The fourth processing module 204 is used to: construct a spatial relationship quantification model for evaluating spatial layout performance based on the semantic association information carried by the digital twin knowledge graph of the medical building; The fifth processing module 205 is used to: generate multiple spatial layout schemes based on the spatial relationship quantification model under preset constraints; The sixth processing module 206 is used to: evaluate the performance of each spatial layout scheme based on the spatial relationship quantification model, and obtain the spatial layout scheme whose evaluation results meet the preset optimization target from the multiple spatial layout schemes.

[0064] It should be noted that the medical building space layout optimization system based on ontology and knowledge graph provided in this embodiment of the invention is used to execute all the process steps of the building space layout optimization method based on medical building ontology in the above embodiment. The working principles and beneficial effects of the two are one-to-one, so they will not be described again.

[0065] In the description of this application, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Therefore, a feature defined as "first" or "second" may explicitly or implicitly include one or more features. In the description of this application, "multiple" means two or more, unless otherwise explicitly specified.

[0066] In the above embodiments, the descriptions of each embodiment have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions in other embodiments.

[0067] The embodiments, implementation methods, and related technical features of this application can be combined and substituted for each other without conflict. The above are merely preferred embodiments of this application and are not intended to limit this application in any way. Any simple modifications, equivalent changes, and alterations made to the above embodiments based on the technical essence of this application without departing from the scope of the technical solution of this application shall still fall within the scope of the technical solution of this application.

Claims

1. A method for optimizing the architectural space layout based on the physical structure of a medical building, characterized in that, include: Constructing an ontology library for the medical building field; The ontology library for the medical building domain is used to define functional space units, medical business activities, personnel roles, equipment and facilities, and their semantic relationships in medical buildings. Acquire data related to medical buildings; the data related to medical buildings includes building and environmental data as well as medical and operational data. Based on the ontology library in the medical building field, semantic mapping, cleaning, and fusion of the medical building-related data are performed to form a digital twin knowledge graph of medical buildings; Based on the semantic association information carried by the digital twin knowledge graph of the medical building, a spatial relationship quantification model is constructed to evaluate the performance of spatial layout. Based on the spatial relationship quantification model, multiple spatial layout schemes are generated under preset constraints. Based on the spatial relationship quantification model, the performance of each spatial layout scheme is evaluated, and the spatial layout scheme that meets the preset optimization target is obtained from the multiple spatial layout schemes.

2. The method according to claim 1, characterized in that, Based on the semantic association information carried by the digital twin knowledge graph of the medical building, a spatial relationship quantification model for evaluating spatial layout performance is constructed, including: Operational data representing the collaborative relationships between different functional units are extracted from the digital twin knowledge graph of the medical building; the operational data includes the frequency of personnel movement between different functional units and the frequency of connection of medical business processes between functional units; Calculate the functional association strength between any two functional units based on the operational data and generate a functional association strength matrix; A digital building model for simulation analysis is obtained from the building and environmental data; the digital twin knowledge graph of the medical building is associated with the digital building model. A multi-agent simulation environment is deployed in the digital building model; the movement of patients, medical staff and supplies within the building space is simulated according to a preset spatial layout scheme. Based on the simulation results, at least one of the spatiotemporal performance parameters is obtained from the flow efficiency index, average travel distance, or spatial congestion point distribution. Using the functional correlation strength matrix and the spatiotemporal performance parameters as inputs, a spatial relationship quantification model is constructed to comprehensively evaluate the merits and demerits of spatial layout.

3. The method according to claim 2, characterized in that, Based on the operational data, calculate the functional association strength between any two functional units and generate a functional association strength matrix, including: For each pair of functional units, obtain the corresponding personnel flow frequency value and medical business process connection frequency value; The personnel mobility frequency and medical business process connection frequency values ​​are normalized to obtain normalized personnel mobility intensity and normalized business connection intensity values. A first weighting coefficient is configured for the normalized personnel mobility intensity value, and a second weighting coefficient is configured for the normalized business connection intensity value; the sum of the first weighting coefficient and the second weighting coefficient is 1. Multiply the first weighting coefficient by the normalized personnel mobility intensity value and add the product of the second weighting coefficient and the normalized business connection intensity value to obtain the comprehensive functional association intensity value of the corresponding functional unit pair. Traverse all functional unit pairs, using each functional unit as the row index and column index; fill the corresponding comprehensive functional association strength value into the corresponding position in the matrix to generate the functional association strength matrix.

4. The method according to claim 3, characterized in that, Associating the digital twin knowledge graph of the medical building with the digital building model includes: Extract spatial objects with independent functional attributes from the digital building model; obtain the unique geometric identifier and spatial semantic label for each spatial object; Semantic definitions of each functional unit are extracted from the digital twin knowledge graph of the medical building; the semantic definitions include at least one of function type, department, and service role. Based on the matching of the spatial semantic tags with the semantic definitions of the functional units, a one-to-one correspondence between spatial objects and functional units is established. The one-to-one correspondence is stored in the medical building digital twin knowledge graph as a link attribute, so that any functional unit can be associated with a specific spatial geometric location in the digital building model.

5. The method according to claim 4, characterized in that, Based on a pre-designed spatial layout, the movement of patients, medical staff, and supplies within the building space is simulated, including: Based on the preset spatial layout scheme, the specific location of each functional unit in the building space is determined; The arrangement locations are mapped to the digital building model to generate a simulation scene corresponding to the spatial layout scheme; In the simulation scenario, intelligent agents representing patients, medical staff, and supplies are configured, and their behavioral rules are set. The agent is driven to run in the simulation scenario, simulating its movement process under the spatial layout scheme.

6. The method according to claim 5, characterized in that, Based on the aforementioned ontology library for the medical building domain, semantic mapping, cleaning, and fusion of relevant data on medical buildings are performed to form a digital twin knowledge graph for medical buildings, including: Geometric information and spatial attributes are extracted from the building and environmental data; the geometric information includes room number, area, and coordinates; the spatial attributes include functional zoning and floor distribution. Business process information and personnel role information are extracted from the medical and operational data; the business process information includes the diagnosis and treatment process and equipment usage records; the personnel role information includes the responsibilities of medical staff and the patient's medical treatment path. Based on the definitions in the ontology library of the medical building domain, the geometric information, spatial attributes, business process information and personnel role information are semantically mapped to generate a unified data structure. The unified data structure is cleaned to remove duplicate, erroneous, or inconsistent data. Based on the predefined semantic relationships in the ontology library of the medical building domain, the cleaned semantic entities are associated; semantic links are established between functional space units, medical business activities, personnel roles and equipment facilities to form knowledge graph nodes and edges; The knowledge graph nodes and edges are stored in a graph database to construct a digital twin knowledge graph for medical buildings.

7. The method according to claim 6, characterized in that, Using the functional correlation strength matrix and the spatiotemporal performance parameters as inputs, a spatial relationship quantification model is constructed to comprehensively evaluate the merits of spatial layout, including: Determine the geometric position of each functional unit in the digital building model according to the preset spatial layout scheme; calculate the spatial distance between any two functional units and generate a spatial distance matrix; The functional association strength matrix and the spatial distance matrix are multiplied element by element to obtain the product matrix of association strength and spatial distance. The layout association cost value is obtained by summing all off-diagonal elements in the product matrix of the association strength and spatial distance. The layout-related cost value and the spatiotemporal performance parameters are fused to form a comprehensive evaluation index; wherein, the fusion method includes using the spatiotemporal performance parameters as hard constraints, accepting the layout scheme only when a preset threshold is met; or, configuring weight coefficients for the layout-related cost value and each spatiotemporal performance parameter respectively, and calculating a weighted comprehensive score; The comprehensive evaluation index is used as the output of the spatial relationship quantification model.

8. The method according to claim 7, characterized in that, Based on the aforementioned spatial relationship quantification model, multiple spatial layout schemes are generated under preset constraints, including: S801. Extract the set of functional units to be laid out from the digital twin knowledge graph of the medical building and obtain the functional association strength between any two functional units. S802. Determine the space occupancy parameters for each functional unit based on the area requirements of each functional unit and the total available space in the building. S803. With the goal of reducing the sum of the product of functional correlation strength and spatial distance, and combined with preset constraints, an initial spatial layout scheme is generated. S804. Input the initial spatial layout scheme into the spatial relationship quantification model to obtain the corresponding layout association cost value and spatiotemporal performance parameters; integrate the layout association cost value and the spatiotemporal performance parameters to form a comprehensive evaluation index; S805. If the comprehensive evaluation index does not meet the preset diversity or convergence conditions, the current layout scheme is perturbed to generate a new candidate layout scheme and the process is returned to S804 for re-evaluation. Repeat steps S804 and S805 until multiple spatial layout schemes that meet the preset constraints and have different comprehensive evaluation indicators are obtained.

9. The method according to claim 8, characterized in that, Based on the aforementioned spatial relationship quantification model, performance evaluation is performed on each spatial layout scheme. From these multiple spatial layout schemes, spatial layout schemes whose evaluation results satisfy the preset optimization objective are obtained, including: The spatial relationship quantification model is invoked to evaluate the multiple spatial layout schemes respectively, and a comprehensive evaluation index corresponding to each scheme is obtained; The comprehensive evaluation indicators of the multiple spatial layout schemes are sorted or filtered according to the preset optimization objectives; wherein, the preset optimization objectives include at least one of the following: minimizing the layout-related cost value, maximizing circulation efficiency, and ensuring that all spatiotemporal performance parameters meet the corresponding preset thresholds; The spatial layout scheme with the best comprehensive evaluation index and that meets all preset constraints is selected as the final optimization result.

10. A medical building spatial layout optimization system based on ontology and knowledge graph, characterized in that, include: The first processing module is used to: build an ontology library for the medical building field; The ontology library for the medical building domain is used to define functional space units, medical business activities, personnel roles, equipment and facilities, and their semantic relationships in medical buildings. The second processing module is used to: acquire data related to medical buildings; the data related to medical buildings includes building and environmental data as well as medical and operational data. The third processing module is used to: perform semantic mapping, cleaning and fusion of the medical building-related data based on the medical building ontology library to form a digital twin knowledge graph of medical buildings; The fourth processing module is used to: construct a spatial relationship quantification model for evaluating spatial layout performance based on the semantic association information carried by the digital twin knowledge graph of the medical building; The fifth processing module is used to: generate multiple spatial layout schemes based on the spatial relationship quantification model, under the condition of satisfying preset constraints; The sixth processing module is used to: evaluate the performance of each spatial layout scheme based on the spatial relationship quantification model, and obtain the spatial layout scheme whose evaluation results meet the preset optimization target from the multiple spatial layout schemes.

Images