First-aid equipment site selection method and system, storage medium and equipment

Abstract

The embodiment of the invention discloses a site selection method and system for first-aid equipment, a storage medium and equipment, and is applied to the technical field of information processing. The site selection system of the first-aid equipment performs gridding processing on a target area to obtain a plurality of grid units, and obtains grid data of each grid unit, including current first-aid equipment, traffic, weather, society and historical out-of-hospital emergency attack events. On the basis of the acquired grid characteristics of each grid unit, spatial connection characteristics among the grid units and time sequence characteristics, after the acquired characteristics are fused, occurrence information of extrahospital emergency attack events corresponding to each grid unit is predicted according to the fused characteristics; therefore, factors of multiple dimensions such as time, space, weather, traffic, society and the like can be comprehensively considered to perform event prediction, and on the basis, the deployment information of the first-aid equipment in the target area is determined in combination with the coverage rate model of the first-aid equipment, so that the first-aid equipment can be planned more reasonably and accurately.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of information processing, in particular to an emergency equipment site selection method and system, a storage medium and equipment. BACKGROUND

[0002] An automated external defibrillator (AED) is an important device for rescuing patients with cardiac arrest. Deploying some places in the city can achieve first aid for out-of-hospital cardiac arrest (OHCA). This requires the AED device to be deployed in places where OHCA events occur more frequently in the city.

[0003] The existing AED device site selection method is mainly based on historical emergency data, population density, medical resource distribution and other factors to select sites. Specifically, based on historical OHCA event data, the event density is calculated and hotspot analysis is performed using geographic information system (GIS) tools, and the high incidence area of OHCA events is identified, and AED site selection is performed by maximizing the coverage population. SUMMARY

[0004] The embodiments of the present application provide an emergency equipment site selection method, system, storage medium and equipment, which improves the rationality and accuracy of the deployment of emergency equipment.

[0005] The embodiments of the present application provide an emergency equipment site selection method, which comprises: Obtaining multi-dimensional data of a target area; the multi-dimensional data comprises current emergency equipment data, traffic data, weather data, social data and historical out-of-hospital emergency event data; Grid processing is performed on the target area to obtain a plurality of grid units, and grid data of each grid unit is obtained according to the multi-dimensional data, wherein the grid data comprises current emergency equipment data, traffic data, weather data, social data and historical out-of-hospital emergency event data of the corresponding grid unit; According to the grid data, the grid characteristics of each grid unit are obtained, and the spatial connection characteristics between the plurality of grid units and the time sequence characteristics of the plurality of grid units are obtained; The grid characteristics, spatial connection characteristics and time sequence characteristics are fused to obtain fused features, and the occurrence information of the out-of-hospital emergency event corresponding to each grid unit is predicted according to the fused features; According to the occurrence information of the out-of-hospital emergency event and the preset coverage rate model of the emergency equipment, the deployment information of the emergency equipment in the target area is determined.

[0006] In another aspect, the embodiment of the present application provides a site selection system of emergency equipment, comprising: a data acquisition unit configured to acquire multi-dimensional data of a target area, wherein the multi-dimensional data comprises current emergency equipment data, traffic data, weather data, social data and historical out-of-hospital emergency event data; a meshing unit configured to mesh the target area to obtain a plurality of mesh units, and acquire mesh data of each mesh unit according to the multi-dimensional data, wherein the mesh data comprises current emergency equipment data, traffic data, weather data, social data and historical out-of-hospital emergency event data of the corresponding mesh unit; a feature acquisition unit configured to acquire mesh features of the mesh units according to the mesh data, and acquire spatial connection features between the mesh units and time sequence features of the mesh units; a prediction unit configured to fuse the mesh features, the spatial connection features and the time sequence features to obtain fused features, and predict occurrence information of out-of-hospital emergency events corresponding to each mesh unit according to the fused features; a deployment unit configured to determine deployment information of the emergency equipment in the target area according to the occurrence information of the out-of-hospital emergency events and a preset coverage rate model of the emergency equipment.

[0007] In another aspect, the embodiment of the present application also provides a computer readable storage medium storing a plurality of computer programs, wherein the computer programs are adapted to be loaded and executed by a processor to implement the site selection method of the emergency equipment according to the first aspect of the embodiment of the present application.

[0008] In another aspect, the embodiment of the present application also provides a terminal device comprising a processor and a memory. The memory is configured to store a plurality of computer programs, wherein the computer programs are used to load and execute the site selection method of the emergency equipment according to the first aspect of the embodiment of the present application by the processor; and the processor is configured to implement each computer program in the plurality of computer programs.

[0009] It can be seen that in the method of the embodiment, the site selection system of the emergency equipment performs grid processing on the target area to obtain a plurality of grid units, and obtains multi-dimensional grid data of each grid unit, and based on this, obtains grid features of each grid unit, spatial connection features between the grid units and time sequence features, fuses the obtained features, and predicts occurrence information of the out-of-hospital emergency event corresponding to each grid unit according to the fused features, and then determines the deployment information of the emergency equipment in the target area in combination with the coverage rate model of the emergency equipment. In this process, time, space, weather, traffic, society and other factors in multiple dimensions are comprehensively considered to predict the out-of-hospital emergency event, and the emergency equipment is planned on this basis, so that the emergency equipment can be more reasonably and accurately planned. BRIEF DESCRIPTION OF DRAWINGS

[0010] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings needed to be used in the embodiments or the prior art description. Obviously, the drawings in the following description are only some embodiments of the present application, and for those skilled in the art, other drawings can also be obtained without creative labor.

[0011] Figure 1 is a schematic diagram of a site selection method of emergency equipment provided by an embodiment of the present application; Figure 2 is a flowchart of a site selection method of emergency equipment provided by a specific embodiment of the present application; Figure 3a is a schematic diagram of a risk prediction model in a specific embodiment of the present application; Figure 3b is another schematic diagram of a wind direction prediction model in a specific embodiment of the present application; Figure 4 is a flowchart of a site selection method of emergency equipment provided by an application embodiment of the present application; Figure 5 is a logical structure schematic diagram of a site selection system of emergency equipment provided by an embodiment of the present application; Figure 6 is a logical structure schematic diagram of a terminal device provided by an embodiment of the present application. DETAILED DESCRIPTION

[0012] The technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only some embodiments of the present application, not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor are within the scope of protection of the present application.

[0013] The terms "first", "second", "third", "fourth" and the like in the description and claims of the present application, and above accompanying drawings, if any, are used as identifiers to distinguish between similar objects, and are not necessarily intended to denote a specific sequential or chronological order. It will be understood that the use of such terms is intended to cover the embodiments of the present application whether used in a dependent or an independent manner. It is to be understood that the data so used is merely illustrative and not restrictive; the scope of the application should not be limited thereto since modifications can become apparent to those skilled in the art, with the benefit of the present disclosure, without departing from the scope of the application, and it is intended to cover such modifications as would fall within the scope of the applicant's contribution to the art. Furthermore, the terms "comprise", "comprising", "include", "including", and "has", "having" and variants thereof are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises, has or includes an item or list of items that does not also preclude the occurrence of additional, un-recited items.

[0014] The embodiments of the present application provide a method for site selection of emergency equipment, which is mainly used for site selection when deploying emergency equipment (such as AED) in a target area, as shown in the following Figure 1 The site selection of emergency equipment such as AED is realized by the site selection system of emergency equipment according to the following method: obtaining multi-dimensional data of a target area; performing grid processing on the target area to obtain a plurality of grid units, and obtaining grid data of each grid unit according to the multi-dimensional data; obtaining grid features of each grid unit according to the grid data, and obtaining spatial connection features between the plurality of grid units and time sequence features of the plurality of grid units; fusing the grid features, the spatial connection features and the time sequence features to obtain fused features, predicting occurrence information of an out-of-hospital emergency event corresponding to each grid unit according to the fused features; and determining deployment information of the emergency equipment in the target area according to the occurrence information of the out-of-hospital emergency event and a preset coverage rate model of the emergency equipment.

[0015] In this way, the target area can be analyzed in detail by taking the grid unit as the analysis unit, the occurrence information of the out-of-hospital emergency event can be accurately predicted by combining the space-time features of the grid unit, and the accuracy of the deployment of the emergency equipment in each grid unit can be improved.

[0016] The embodiments of the present application provide a method for site selection of emergency equipment, which is mainly a method executed by a site selection system of emergency equipment, as shown in the following Figure 2 The method comprises the following steps: Step 101, obtaining multi-dimensional data of a target area.

[0017] It can be understood that when the flow of the present embodiment is initiated, the following multi-dimensional data can be obtained, but is not limited to: current emergency equipment data, traffic data, weather data, social data, and the like, and can also include historical out-of-hospital emergency event data, and the like, wherein: The historical out-of-hospital acute illness episode data can be obtained from emergency service agencies, including information such as episode time, location, medical emergency response time, underlying disease, and the like.

[0018] The current emergency equipment data can include the latitude and longitude, address information, attributes, and state information of the current emergency equipment, as well as potential emergency equipment point of interest (POI) data. The emergency equipment here mainly refers to some emergency equipment deployed for on-site emergency assistance for out-of-hospital acute illness, such as AED equipment and the like.

[0019] The traffic data of the target area can be obtained through an intelligent transportation system, including traffic flow and road congestion conditions and the like.

[0020] The weather data of the target area can be obtained from meteorological departments, including historical weather conditions such as wind, rain, and snow and the like.

[0021] The social data of the target area: administrative division, POI data, real-time population distribution, and population statistics such as population density and the like.

[0022] In step 102, the target area is subjected to grid processing to obtain a plurality of grid units, and grid data of each grid unit is obtained according to the multi-dimensional data, including current emergency equipment data, traffic data, weather data, social data, and historical out-of-hospital acute illness episode data of the corresponding grid unit.

[0023] Specifically, the target area can be subjected to grid processing using a city information modeling (CIM) platform, and each grid unit represents a small geographic area. The size of each grid unit can be specified according to the size, data density, and actual requirements of the target area, and the plurality of grid units obtained need to cover all geographic points of the target area, such as 100m x 100m or 200m x 200m.

[0024] The CIM is based on building information modeling (BIM), geographic information system (GIS), Internet of Things (IoT), and the like, integrates multi-dimensional and multi-scale spatial data and Internet of Things sensing data of cities above and below ground, indoors and outdoors, and the past, present, and future, and constructs a three-dimensional digital space city information organic complex.

[0025] Further, when obtaining the grid data of each grid unit, a unique identification (ID) can be set for each grid unit, and the population density, traffic flow, and number of historical out-of-hospital acute illness episode events and the like of each grid unit are obtained based on the above multi-dimensional data, achieving the grid of the above multi-dimensional data.

[0026] It should be noted that before the gridding of multi-dimensional data, the multi-dimensional data can also be pre-processed, which can specifically include: Data cleaning can include but is not limited to the following data cleaning processing: missing value processing and removing outliers; The multi-dimensional data is subjected to multi-dimensional normalization processing, such as time normalization, space normalization and time alignment, to obtain normalized data.

[0027] In this way, when the grid data of each grid cell is obtained, the multi-dimensional standardized data corresponding to each grid cell can be obtained according to the normalized data.

[0028] Specifically, in the missing value processing, interpolation method, average method, etc. can be used to fill in the data for the missing values. For example, if the traffic flow data at a certain time point is missing, the average value of the traffic flow at the previous and subsequent time points can be used to fill in.

[0029] In removing outliers, outliers (such as impossible high flow data) can be identified through statistical analysis and visualization means, and corrected or deleted according to reasonable rules.

[0030] In time normalization processing, the timestamps of all data can be unified to the same time zone and converted to a unified time format (such as Unix timestamp).

[0031] In space normalization processing, addresses can be converted to CGCS2000 unified longitude and latitude coordinates (geocoding) for subsequent spatial analysis.

[0032] In time alignment processing, in order to fuse multi-dimensional data, alignment needs to be performed according to the timestamp. For example, traffic data and out-of-hospital acute event data in the same time period are aligned.

[0033] Step 103, obtaining the grid features of each grid cell according to the grid data obtained in step 102, and obtaining the spatial connection features between the plurality of grid cells and the time sequence features of the plurality of grid cells.

[0034] Specifically, the obtained grid features can be a spatial feature matrix, the spatial connection features can be a spatial adjacency matrix, and the time sequence features refer to the feature values of each grid cell at different time instants. Among them: In the process of obtaining the grid features, the static features and dynamic features can be obtained according to the grid data obtained above. The static features can include the population flow, population density, age, distance to medical institutions, actual geographical location, historical out-of-hospital acute illness event data such as emergency response time, cardiac patient population distribution, hypertension population distribution, diabetes population distribution, existing emergency equipment distribution, emergency equipment area opening time, average income level of personnel, and the like. The dynamic features can include real-time traffic flow and current weather conditions of the grid unit.

[0035] Specifically, the static features and dynamic features of any grid unit can be normalized, and then the normalized static features and dynamic features are weighted and averaged to obtain the grid features of the grid unit.

[0036] In the process of obtaining the spatial connection features, the connection relationship between the grid units can be established, and the weight of the connection between the grid units can be obtained to identify the strength of the mutual influence between the grid units. Specifically, the connection relationship between the grid units can be calculated based on the geographical distance. The closer the distance, the greater the weight, and the weight can be dynamically adjusted according to the traffic flow, population flow, and the like.

[0037] Among them, the edge index matrix edge_index can be obtained according to the connection relationship between the grid units, and the edge weight matrix edge_weights can be obtained according to the weight of the connection between the grid units.

[0038] In the process of obtaining the time sequence features, different time periods can be constructed. Specifically, a week can be divided into two groups of weekdays and weekends, and each group can be further divided into multiple time periods according to a small time interval. Then, the grid features and spatial connection features of the grid units in the multiple time periods can be collected to obtain the time sequence features.

[0039] Specifically, an array for representing the deployment of emergency equipment can be obtained according to the time sequence features. Each element in the array can represent whether there is available emergency equipment in a grid unit in a time period. If there is, the element value is 1, and if there is not, the element value is 0. It should be noted that whether the emergency equipment is available mainly includes two aspects: one is the service state of the emergency equipment, which is normal or abnormal, and the other is the service state of the place where the emergency equipment is located, which considers the working hours and business hours. Only when the service state of the emergency equipment is normal and the service state of the place where the emergency equipment is located is business, the element value in the corresponding array is 1.

[0040] In step 104, the grid features, spatial connection features, and time sequence features are fused to obtain fused features, and the occurrence information of the out-of-hospital acute illness event corresponding to each grid unit is predicted according to the fused features.

[0041] Specifically, a preset risk prediction model of an out-of-hospital acute event can be invoked, the grid features, the spatial connection features and the time sequence features are fused by the risk prediction model to obtain fused features, and the occurrence probability of the out-of-hospital acute event corresponding to each grid unit, i.e., the occurrence probability of the out-of-hospital acute event, is predicted based on the fused features.

[0042] The risk prediction model is a machine learning model based on artificial intelligence. Artificial intelligence (AI) is the theory, method, technology and application system for using digital computers or digital computer controlled machines to simulate, extend and expand human intelligence, perceive environment, acquire knowledge and use knowledge to obtain the best results. In other words, artificial intelligence is a comprehensive technology of computer science, which attempts to understand the essence of intelligence and produce a new intelligent machine that can react in a similar way to human intelligence. Artificial intelligence is to study the design principles and implementation methods of various intelligent machines, so that machines have the functions of perception, reasoning and decision-making.

[0043] Artificial intelligence technology is a comprehensive discipline, involving a wide range of fields, both hardware and software technologies. Artificial intelligence basic technologies generally include sensors, special artificial intelligence chips, cloud computing, distributed storage, big data processing technology, operation / interaction system, mechatronics and other technologies. Artificial intelligence software technology mainly includes computer vision technology, speech processing technology, natural language processing technology, machine learning and deep learning.

[0044] Machine learning (ML) is a multi-disciplinary subject involving probability theory, statistics, approximation theory, convex analysis, algorithm complexity theory and other disciplines. It is a subject that studies how computers simulate or implement human learning behavior to acquire new knowledge or skills, and reorganize existing knowledge structure to continuously improve their performance. Machine learning is the core of artificial intelligence and the fundamental approach to making computers intelligent, and its applications are widespread in various fields of artificial intelligence. Machine learning and deep learning usually include artificial neural networks, belief networks, reinforcement learning, transfer learning, inductive learning, and teaching learning.

[0045] In the specific example process, as shown in Figure 3a The risk prediction model can include a spatial feature layer 10, a space-time sequence layer 11 and an output layer 12, wherein: The spatial feature layer 10 is configured to perform convolution processing on the grid features and the spatial connection features of any grid unit to obtain spatial convolution features; The spatio-temporal sequence layer 11 is configured to fuse the time sequence features and the spatial convolution features of the grid cell to obtain fused features. The output layer 12 is configured to output the occurrence probability of the out-of-hospital acute illness onset event of the grid cell based on the fused features. For example, when the occurrence probability exceeds 0.4, the grid cell should be equipped with an emergency device, and when the occurrence probability exceeds 0.7, the grid cell is considered as a high-risk area and needs to be equipped with multiple emergency devices.

[0046] Further, as shown in Figure 3b The risk prediction model can further include an attention layer 13 configured to process the spatial convolution features by using an attention mechanism to obtain spatial attention features, that is, some important features in the spatial convolution features are focused on and enhanced, and some unimportant features are weakened. Then, the spatio-temporal sequence layer 11 fuses the spatial attention features and the time sequence features to obtain the fused features.

[0047] In the implementation process, the spatial feature layer 10 can be a spatio-temporal graph neural network (ST-GNN) specially used for processing spatio-temporal data such as traffic flow prediction, weather forecasting, and social network analysis. The ST-GNN combines the characteristics of a graph neural network (GNN) and a recurrent neural network (RNN) and can capture the spatial dependence and temporal dynamics of the data.

[0048] The spatio-temporal sequence layer 11 can be a bidirectional long short-term memory (LSTM) network or a gated recurrent unit (GRU) configured to process the time sequence correlation of the spatio-temporal features and enhance the modeling capability of the time sequence data.

[0049] It should be noted that one risk prediction model can correspond to each grid cell, and multiple risk prediction models can be processed in parallel to predict the occurrence probability of the out-of-hospital acute illness onset event of the corresponding grid cell. Thus, the prediction efficiency of the out-of-hospital acute illness onset event of multiple grid cells is improved.

[0050] In step 105, the deployment information of the emergency device in the target area is determined according to the occurrence information of the out-of-hospital acute illness onset event and the preset coverage model of the emergency device.

[0051] It should be noted that the coverage model of the emergency equipment is mainly based on the principle of minimizing the emergency response time or maximizing the high-risk area, and the location planning of the emergency equipment is carried out to realize the reasonable deployment and efficient emergency of the AED equipment. The coverage model can select a multi-objective optimization algorithm such as a multi-objective particle swarm optimization algorithm (MOPSO) and a genetic algorithm (GA) for optimization and solution, and obtain the deployment information of the emergency equipment in some grid cells in the target area. Specifically: According to the occurrence information of the out-of-hospital emergency event, the potential grid cells for deployment of the emergency equipment are set, for example, areas with large flow of people such as shopping malls, schools, bus stations, supermarkets, and grid cells with an occurrence probability of out-of-hospital emergency events exceeding a set value; Based on the coverage model of the maximum risk area criterion, the service radius of the emergency equipment in the potential grid cell is set as R; or based on the coverage model of the minimum emergency response time criterion, the deployment information of the emergency equipment in the potential grid cell is set.

[0052] Among them, the service radius obtained by iteration of different site selection algorithms can be selected, and the site selection algorithm whose calculated service radius can cover the area with a higher probability of occurrence of out-of-hospital emergency events is selected, and the service radius is obtained based on the selected site selection algorithm. Each site selection algorithm can include: based on a multi-objective optimization algorithm such as a multi-objective particle swarm optimization algorithm (MOPSO) and a genetic algorithm (GA) for optimization and solution, which can ensure that the high-risk area (area with a higher probability of occurrence of out-of-hospital emergency events) can quickly obtain emergency services.

[0053] Or based on the coverage model of the minimum emergency response time criterion, the shortest time cost of reaching the deployment point of the emergency equipment in the grid cell under different site selection algorithms is calculated, the site selection algorithm of the AED equipment with the shortest time cost is selected, and the deployment information of the emergency equipment is obtained based on the selected site selection algorithm, so as to improve the success rate of emergency.

[0054] Specifically, taking the multi-objective particle swarm optimization algorithm as an example, specifically: Initialize the particle swarm, which includes a plurality of particles in the particle swarm, each particle corresponds to a grid cell, and each particle is randomly assigned an initial position and velocity; Calculate the fitness value of each particle, compare and store the non-inferior solution of the optimal solution group by using the Pareto front method; Update the position and velocity of each particle: Update the particle's own optimal solution and global optimal solution according to the fitness function, select the final solution according to the actual needs, and generate the deployment information of the emergency equipment in each grid cell according to the optimization algorithm result.

[0055] Further, the determined deployment information of the emergency equipment in the grid unit can be visually displayed through the CIM platform.

[0056] It can be seen that in the method of the embodiment, the site selection system of the emergency equipment performs grid processing on the target area to obtain a plurality of grid units, and obtains grid data of each grid unit. Based on this, the grid features of each grid unit, the spatial connection features between the grid units, and the time sequence features are obtained. After fusing the obtained features, the occurrence information of the out-of-hospital emergency event corresponding to each grid unit is predicted according to the fused features. Then, in combination with the coverage rate model of the emergency equipment, the deployment information of the emergency equipment in the target area is determined. In this process, factors in multiple dimensions such as time, space, weather, traffic, and society are comprehensively considered for the prediction of the out-of-hospital emergency event. On this basis, the emergency equipment is planned, and the emergency equipment can be more reasonably and accurately planned.

[0057] It should be noted that, in order to realize the prediction of the occurrence information of the out-of-hospital emergency event corresponding to each grid unit in the above embodiment, a risk prediction model can be trained in advance. Specifically, when training the risk prediction model: a risk prediction original model is constructed. The risk prediction original model can include a spatial feature layer, a spatio-temporal sequence layer, and an output layer as shown in the above Figure 3a , or a structure as shown in the above Figure 3b ; training samples are obtained. The training samples include: grid features, spatial connection features, and time sequence features of a plurality of sample grids in a sample area, and information of historical occurrence of out-of-hospital emergency events of each sample grid; the risk prediction original model is trained according to the training samples to obtain the above-mentioned preset risk prediction model.

[0058] Wherein, the method for obtaining the grid features, spatial connection features, and time sequence features of the plurality of sample grids in the sample area included in the training samples is similar to the method for obtaining the grid features, spatial connection features, and time sequence features of each grid unit in the target area, and will not be described here.

[0059] Specifically, when training the risk prediction original model, the grid features, spatial connection features, and time sequence features of each sample grid can be fused by the risk prediction original model, and the probability of occurrence of the out-of-hospital emergency event of each sample grid is predicted according to the fused features. A loss function, such as a binary cross-entropy loss function, is set to measure the error between the probability predicted by the risk prediction original model and the information of historical occurrence of out-of-hospital emergency events of each sample grid. Then, a gradient is calculated based on the value of the loss function, and the Adam optimization algorithm is used for parameter optimization. The learning rate and the regularization parameter can be adjusted to prevent overfitting.

[0060] Further, the K-fold cross-validation method can be used to adjust the hyperparameters in the risk prediction original model, such as learning rate, convolution kernel size, number of graph convolution layers, number of time convolution layers, etc.

[0061] Through multiple adjustments of the parameters in the risk prediction original model, the loss function meets certain conditions, and at this time a risk prediction model is obtained, which can be preloaded into the system to execute the above step 104 when initiating the site selection process of the emergency equipment.

[0062] The following specific application example illustrates the site selection method of the emergency equipment of the embodiment. In the embodiment, the emergency equipment is AED equipment, and the out-of-hospital acute illness event is an OHCA event. As shown in Figure 4 , it includes: Step 201, obtaining multi-dimensional data of a sample area, which can include historical OHCA event data, AED data, traffic data, weather data, and social data, etc.

[0063] Step 202, performing grid processing on the sample area to obtain a plurality of sample grids, and obtaining grid data of each sample grid according to the multi-dimensional data.

[0064] Step 203, obtaining training samples, which include: obtaining grid features of each sample grid according to the grid data obtained in step 202, obtaining spatial connection features between a plurality of sample grids, obtaining time sequence features of a plurality of sample grids, and obtaining information of historical OHCA events of each sample grid.

[0065] Step 204, constructing a risk prediction original model, the structure of which can be as shown in Figure 3a or Figure 3b .

[0066] Step 205, fusing the grid features of each sample grid, the spatial connection features between a plurality of sample grids, and the time sequence data through the risk prediction original model to obtain fused features, predicting the occurrence probability of the OHCA event corresponding to each sample grid based on the fused features, and then setting a loss function according to the occurrence probability of the OHCA event corresponding to the sample grid and the information of the historical OHCA event of the corresponding sample grid in the training sample, and optimizing the hyperparameters in the risk prediction original model based on the loss function to obtain a trained risk prediction model.

[0067] After training the risk prediction model, the risk prediction model is preloaded into the site selection system of the AED equipment. When initiating the site selection process of the AED equipment in the target area, the following steps 206 to 209 can be executed.

[0068] In step 206, multi-dimensional data of the target area is acquired, which can include historical OHCA event data, AED data, traffic data, weather data, and social data, etc.

[0069] In step 207, the target area is grid processed to obtain a plurality of grid cells, and grid data of each grid cell is acquired according to the multi-dimensional data, and then grid features of each grid cell are acquired according to the grid data, and spatial connection features between the plurality of grid cells are acquired, and time sequence features of the plurality of grid cells are acquired.

[0070] In step 208, a preset risk prediction model of OHCA events is called, the grid features, the spatial connection features, and the time sequence features are fused by the risk prediction model to obtain fused features, and occurrence probabilities of OHCA events corresponding to each grid cell are predicted based on the fused features.

[0071] In step 209, deployment information of AED devices in each of the target areas is determined according to the occurrence probabilities of the OHCA events and a preset coverage model of the AED devices.

[0072] It can be seen that the AED device site selection method of the embodiment can achieve the following technical effects: In the risk prediction model, the time sequence features and the spatial features are introduced into the graph neural network at the same time to form a space-time graph structure, so that the model can capture the complex interaction relationship of the OHCA events in time and space, and can better predict the dynamic changing risk.

[0073] High-precision OHCA event risk prediction: through the risk prediction model, data analysis is performed in two dimensions of time and space, which can more accurately predict OHCA high-risk areas and generate detailed risk heat maps, so that planning personnel can intuitively understand the OHCA risk distribution of each area.

[0074] In the AED device site selection process, a plurality of optimization objectives such as response time, coverage rate, and passenger flow are comprehensively considered, a minimum first-aid reaction time is proposed according to actual application scenarios and environmental changes, a maximum coverage rate model of high-risk areas is proposed, and an optimal facility layout scheme is determined, so that the practicability and flexibility of the site selection scheme are ensured.

[0075] The embodiment of the application further provides an emergency device site selection system, a structure diagram of which is shown in Figure 5 As shown in the figure, the system can specifically include: A data acquisition unit 20 is configured to acquire multi-dimensional data of a target area. The multi-dimensional data includes current emergency device data, traffic data, weather data, social data, and historical out-of-hospital acute illness event data.

[0076] The grid unit 21 is configured to grid the target area to obtain a plurality of grid units, and obtain grid data of each grid unit according to the multidimensional data obtained by the data acquisition unit 20. The grid data includes current emergency equipment data, traffic data, weather data, social data, and historical out-of-hospital acute event data of the corresponding grid unit.

[0077] The grid unit 21 is configured to perform time normalization, space normalization, and time alignment on the multidimensional data to obtain normalized data when obtaining the grid data of each grid unit. A unique identifier is set for each grid unit, and the multidimensional standardized data corresponding to each grid unit is obtained according to the normalized data.

[0078] The feature acquisition unit 22 is configured to obtain grid features of each grid unit, spatial connection features between the plurality of grid units, and time sequence features of the plurality of grid units according to the grid data obtained by the grid unit 21.

[0079] The prediction unit 23 is configured to fuse the grid features, spatial connection features, and time sequence features obtained by the feature acquisition unit 22 to obtain fused features, and predict occurrence information of an out-of-hospital acute event corresponding to each grid unit according to the fused features.

[0080] The prediction unit 23 is configured to call a preset risk prediction model of an out-of-hospital acute event, fuse the grid features, spatial connection features, and time sequence features by the risk prediction model to obtain fused features, and predict an occurrence probability of an out-of-hospital acute event corresponding to each grid unit based on the fused features.

[0081] The risk prediction model can include a spatial feature layer, a space-time sequence layer, and an output layer. The spatial feature layer is configured to perform convolution processing on the grid features and spatial connection features of any grid unit to obtain spatial convolution features. The space-time sequence layer is configured to fuse the time sequence features and spatial convolution features to obtain fused features. The output layer is configured to output an occurrence probability of an out-of-hospital acute event of the corresponding grid unit based on the fused features.

[0082] Further, the risk prediction model further includes an attention layer configured to process the spatial convolution features by using an attention mechanism to obtain spatial attention features. The space-time sequence layer is configured to fuse the spatial attention features and time sequence features to obtain fused features.

[0083] The deployment unit 24 is configured to determine deployment information of the emergency equipment in the target area according to the occurrence information of the out-of-hospital emergency event predicted by the prediction unit 23 and the preset coverage model of the emergency equipment.

[0084] The deployment unit 24 is specifically configured to set a potential grid unit of the emergency equipment deployment according to the occurrence information of the out-of-hospital emergency event, set a service radius of the emergency equipment in the potential grid unit based on a coverage model of a maximum risk area criterion, or set the deployment information of the emergency equipment in the potential grid unit based on a coverage model of a minimum emergency response time criterion.

[0085] Further, the site selection system of the emergency equipment in the embodiment can further include: The training unit 25 is configured to construct a risk prediction original model including a spatial feature layer, a spatio-temporal sequence layer and an output layer, acquire a training sample including grid features, spatial connection features and time sequence features of a plurality of sample grids in a sample area and information of historical out-of-hospital emergency events of each sample grid, and train the risk prediction original model according to the training sample to obtain the preset risk prediction model.

[0086] In the system in the embodiment, the factors in multiple dimensions such as time, space, weather, traffic and society are comprehensively considered in the process to predict the out-of-hospital emergency event, and the emergency equipment is planned on this basis, so that the emergency equipment can be more reasonably and accurately planned.

[0087] The embodiment of the application further provides a terminal device, a structure diagram of which is shown in Figure 6 The terminal device can be greatly different due to different configurations or performances, and can include one or more central processing units (CPUs) 30 (for example, one or more processors) and a memory 31, one or more storage media 32 (for example, one or more mass storage devices) storing an application program 321 or data 322. The memory 31 and the storage medium 32 can be temporary storage or persistent storage. The program stored in the storage medium 32 can include one or more modules (not shown in the figure), and each module can include a series of instruction operations in the terminal device. Further, the central processing unit 30 can be configured to communicate with the storage medium 32 and execute a series of instruction operations in the storage medium 32 on the terminal device.

[0088] Specifically, the application program 321 stored in the storage medium 32 comprises an application program of the emergency equipment locating, and the program can comprise the data acquisition unit 20, the gridding unit 21, the feature acquisition unit 22, the prediction unit 23, the deployment unit 24, and the training unit 25 in the emergency equipment locating system as described above, which will not be repeated here. Further, the central processor 30 can be configured to communicate with the storage medium 32, and perform a series of operations corresponding to the application program of the emergency equipment locating stored in the storage medium 32 on the terminal device.

[0089] The terminal device can further comprise one or more power supplies 33, one or more wired or wireless network interfaces 34, one or more input / output interfaces 35, and / or one or more operating systems 323, such as Windows ServerTM, Mac OS XTM, UnixTM, LinuxTM, FreeBSDTM, etc.

[0090] The steps performed by the emergency equipment locating system in the method embodiments described above can be based on the emergency equipment locating system as described above. Figure 6 The structure of the terminal device is shown.

[0091] Further, the embodiments of the present application also provide a computer readable storage medium storing a plurality of computer programs, which are adapted to be loaded and executed by a processor to perform the emergency equipment locating system method performed by the emergency equipment locating system as described above.

[0092] Further, the embodiments of the present application also provide a terminal device comprising a processor and a memory. The memory is configured to store a plurality of computer programs, which are used to be loaded and executed by the processor to perform the emergency equipment locating system method performed by the emergency equipment locating system as described above; and the processor is configured to implement each of the plurality of computer programs.

[0093] Those skilled in the art can understand that all or part of the steps in the various methods of the above embodiments can be completed by instructing the relevant hardware through a program, and the program can be stored in a computer readable storage medium, which can include read-only memory (ROM), random access memory (RAM), magnetic disk or optical disk, etc.

[0094] The above describes in detail the site selection method, system, storage medium and device of the first aid device provided by the embodiment of the present application. The principles and implementation manners of the present application are described by applying specific examples. The above description of the embodiments is only used to help understand the method of the present application and its core idea. Meanwhile, for those skilled in the art, the specific implementation manners and application ranges will be changed according to the idea of the present application. In conclusion, the content of the specification should not be understood as a limitation of the present application.

Claims

1. A method for selecting the location of emergency medical equipment, characterized in that, include: Obtain multi-dimensional data for the target region; The multi-dimensional data includes: current emergency medical equipment data, traffic data, weather data, social data, and historical out-of-hospital emergency events data; The target area will be gridded to obtain multiple grid cells, and grid data of each grid cell will be obtained based on the multi-dimensional data. The grid data includes current emergency medical equipment data, traffic data, weather data, social data and historical out-of-hospital emergency event data of the corresponding grid cell. Based on the grid data, obtain the grid features of each grid cell, the spatial connection features between the multiple grid cells, and the temporal features of the multiple grid cells; The grid features, spatial connectivity features, and temporal features are fused to obtain fused features. Based on the fused features, the occurrence information of outpatient acute illness events corresponding to each grid unit is predicted. Based on the occurrence information of the out-of-hospital emergency events and the pre-set coverage model of the emergency medical equipment, the deployment information of the emergency medical equipment in the target area is determined, wherein the coverage model is...

2. The method as described in claim 1, characterized in that, The step of obtaining the grid data of each grid cell based on the multi-dimensional data specifically includes: The multi-dimensional data is subjected to time normalization, spatial normalization, and time alignment to obtain normalized data; Each grid cell is assigned a unique identifier, and multidimensional standardized data for each grid cell is obtained based on the normalized data.

3. The method as described in claim 1, characterized in that, The process of fusing the grid features, spatial connectivity features, and temporal features to obtain fused features, and predicting the occurrence information of outpatient acute illness events corresponding to each grid unit based on the fused features, specifically includes: Invoke the pre-set risk prediction model for outpatient emergency events; The risk prediction model fuses the grid features, spatial connectivity features, and temporal features to obtain fused features, and predicts the probability of occurrence of out-of-hospital acute illness events corresponding to each grid unit based on the fused features.

4. The method as described in claim 3, characterized in that, The risk prediction model includes: a spatial feature layer, a spatiotemporal sequence layer, and an output layer, wherein: The spatial feature layer is used to perform convolution processing on the grid features and spatial connectivity features of any grid cell to obtain spatial convolution features; The spatiotemporal sequence layer is used to fuse the temporal features and spatial convolutional features to obtain fused features; The output layer is used to output the probability of occurrence of outpatient acute illness events for the corresponding grid cells based on the fused features.

5. The method as described in claim 4, characterized in that, The risk prediction model also includes: The attention layer is used to process the spatial convolutional features using an attention mechanism to obtain spatial attention features; The spatiotemporal sequence layer is used to fuse the spatial attention features and temporal features to obtain fused features.

6. The method according to any one of claims 1 to 5, characterized in that, The method further includes: A risk prediction primitive model is constructed, which includes a spatial feature layer, a spatiotemporal sequence layer, and an output layer. Acquire training samples, which include: grid features, spatial connectivity features and temporal features of multiple sample grids in the sample region, and information on outpatient acute illness events that have occurred in the history of each sample grid; The original risk prediction model is trained based on the training samples to obtain a preset risk prediction model.

7. The method according to any one of claims 1 to 5, characterized in that, The step of determining the deployment information of the emergency medical equipment in the target area based on the occurrence information of the out-of-hospital emergency medical events and the pre-set coverage model of the emergency medical equipment specifically includes: Based on the occurrence information of the out-of-hospital emergency illness events, potential grid units for the deployment of the emergency medical equipment are set; The service radius of emergency medical equipment in the potential grid cell is set based on a coverage model that maximizes the risk area criterion, or the deployment information of emergency medical equipment in the potential grid cell is set based on a coverage model that minimizes the emergency response time criterion.

8. A location selection system for emergency medical equipment, characterized in that, include: The data acquisition unit is used to acquire multi-dimensional data of the target area. The multi-dimensional data includes: current emergency medical equipment data, traffic data, weather data, social data, and historical out-of-hospital emergency events data; A gridded unit is used to perform gridding processing on the target area to obtain multiple grid units, and to obtain grid data of each grid unit based on the multi-dimensional data. The grid data includes current emergency equipment data, traffic data, weather data, social data, and historical out-of-hospital emergency event data of the corresponding grid unit. The feature acquisition unit is used to acquire the grid features of each grid cell, the spatial connection features between the multiple grid cells, and the temporal features of the multiple grid cells based on the grid data. The prediction unit is used to fuse the grid features, spatial connectivity features, and temporal features to obtain fused features, and to predict the occurrence information of outpatient acute illness events corresponding to each grid unit based on the fused features. The deployment unit is used to determine the deployment information of the emergency medical equipment in the target area based on the occurrence information of the out-of-hospital emergency medical event and the coverage model of the pre-set emergency medical equipment.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a plurality of computer programs adapted to be loaded by a processor and executed by the location method for emergency medical devices as described in any one of claims 1 to 7.

10. A terminal device, characterized in that, Including processor and memory; The memory is used to store a plurality of computer programs, which are loaded by a processor and executed as the location method for emergency medical devices as described in any one of claims 1 to 7; the processor is used to implement each of the plurality of computer programs.

Images