Public health management method and system

Abstract

The invention relates to the technical field of medical care information, in particular to a public health management method and system. The management method comprises the following steps: S1, collecting geographic information of a management area, generating a map and performing area division; s2, collecting big data based on personnel information, event information and environment information, and performing regional planning on personnel in the management region; s3, constructing a risk assessment model based on a BP neural network, performing model optimization by adopting a particle swarm algorithm, and assessing a risk level in each region; s4, marking the map according to the risk level to obtain a risk distribution map; and S5, judging whether the risk level of each area exceeds a preset initial risk level or not, if so, giving an alarm, and judging whether each person exceeds the passable area or not, and if so, giving an alarm. According to the invention, the BP neural network optimized by the particle swarm optimization is adopted to carry out risk assessment on the management area, and the method has the characteristics of high accuracy, fast operation speed and good robustness.

Description

Technical Field

[0001] The present invention relates to the field of medical care information technology, and in particular to a public health management method and a public health management system. Background Art

[0002] Public health management refers to the process of systematically managing various factors that affect group health through organization, planning, coordination, execution and supervision to prevent diseases, promote health and improve the quality of life. For the area under the jurisdiction of the hospital, the content of public health management mainly includes: 1. Health record management: establishing, using and maintaining health records for people entering and leaving; 2. Health education: providing health education materials, setting up health education bulletin boards, carrying out health consultation services, holding health knowledge lectures, and conducting individualized health education; 3. Basic health management: disinfecting risk areas, cleaning the ground and handling garbage throughout the area; 4. Personnel entry and exit management: regional planning for patients, accompanying personnel, and medical staff, prohibiting people from entering and leaving at will or staying in the permitted area for a long time.

[0003] The defects in existing public health management mainly include: 1. Basic health management is not timely, and some areas are not disinfected and garbage is not cleaned for a long time, resulting in the virus not being eliminated in time; 2. Personnel supervision relies on the active observation of supervisors and the conscious obedience of people entering and leaving. Some patients and accompanying personnel enter and exit risk areas at will; 3. The risk assessment of each area is inaccurate, and it is difficult to detect risk changes in various areas in time; 4. The management data is complex, and it is difficult to detect the risk level of each area in time, resulting in untimely emergency response. Summary of the invention

[0004] Based on this, it is necessary to propose a public health management method and system to address the problems of existing public health management due to inaccurate and untimely risk assessment, as well as insufficiently precise and timely personnel supervision.

[0005] The present invention is implemented by the following technical solution: A public health management method comprises the following steps:

[0006] S1: Collect geographic information of the management area, generate maps and divide the area.

[0007] S2: Collect big data based on personnel information, event information and environmental information. Carry out regional planning for personnel in the management area.

[0008] S3: Construct a risk assessment model based on BP neural network and use particle swarm algorithm to optimize the model. Input the current data of each area into the risk assessment model to obtain the risk level in each area.

[0009] S4: Mark the map according to the risk level to obtain a risk distribution map.

[0010] S5: Determine whether the risk level of each area exceeds the preset initial risk level, and issue an alarm if it does. Determine whether each person exceeds the passable area, and issue an alarm if it does.

[0011] The above management method uses an optimized BP neural network to conduct risk assessment on each area. When the assessed risk of any area is higher than the initial risk level, a warning is issued in time, and then it is processed as soon as possible to reduce the probability of risk events. The BP neural network optimized by the particle swarm algorithm has the characteristics of high accuracy, fast calculation speed and good robustness, which makes public health management more timely and has more accurate assessment results, thereby maintaining a good public health status and reducing the risk of infection.

[0012] Furthermore, personnel information includes density data and health data. Environmental information includes air data and ground data. Event information includes small event data based on individual infection and large event data based on collective infection.

[0013] Furthermore, the big data is normalized and expressed as:

[0014]

[0015] Where: X′ represents the result after data normalization, R max , R min represents the maximum and minimum values ​​of the normalized interval, X max , X min Represents the maximum and minimum values ​​of the input layer.

[0016] Furthermore, the construction method of the risk assessment model is as follows:

[0017] S31: Construct an initial BP neural network.

[0018] S32: Introduce particle swarm algorithm to optimize BP neural network.

[0019] Further, in step S31, the BP network model includes an input layer, a hidden layer and an output layer.

[0020] The output of the hidden layer is expressed as:

[0021]

[0022] Where: ij is the weight between the input layer and the hidden layer, θ j is the hidden layer threshold, and n is the number of hidden layer nodes.

[0023] The output of the output layer is expressed as:

[0024]

[0025] Where: jk is the weight between the hidden layer and the output layer, θ k is the output layer threshold, and k is the number of output layer nodes.

[0026] Furthermore, in step S32, the particle update strategy is expressed as:

[0027] v id (t+1)=ω d v id (t)+c 1 r 1 [P id (t)-X id (t)]+c 2 r 2 [P gd (t)-X id (t)]

[0028] X id (t+1)=X id (t)+v id (t)

[0029] Where: p id (t), p gd (t) are the individual optimal solution and the global optimal solution at time t, d is the dimension of the current particle, and d = 1, 2, ..., N, ω d is the particle’s inertia weight, c 1 、c 2 is the learning factor when the particle is updated, r 1 、r 2 is a random number uniformly distributed between [0, 1].

[0030] The inertia weight of the particle swarm is adjusted, expressed as:

[0031]

[0032] Where: L is the space length, X id is the coordinate value of the particle, is the average value of all particle coordinates.

[0033] Furthermore, in step S33, the training method of the BP neural network is specifically as follows:

[0034] 1. Determine the fitness function of the particle swarm algorithm. 2. Initialize the parameters of the BP neural network. 3. Calculate the fitness and weight. 4. Calculate the particle speed and position. 5. Optimize the membership function. 6. Update the particle speed and position. 7. Adjust the weight and threshold of the BP neural network. 8. Determine whether the test accuracy is met. If yes, output the global optimal solution. Otherwise, repeat the above steps.

[0035] The present invention also provides a public health management system, which includes a data collection module, a map generation module, a BP neural network model, a marking module and an early warning module.

[0036] The data collection module is used to collect big data based on personnel information, event information, environmental information, and geographic information of the management area.

[0037] The map generation module is used to generate a management area map based on geographic information, and to divide the management area map into regions and mark initial risks.

[0038] The BP neural network model is used to conduct risk assessment on each sub-region within the management area based on big data and the management area map, and output the risk level of each sub-region.

[0039] The marking module is used to mark risks on the management area map according to the risk level.

[0040] The early warning module is used to determine whether the risk level of each area exceeds the preset initial risk level, and to issue an alarm if it does. It is also used to determine whether each person exceeds the passable area, and to issue an alarm if it does.

[0041] Compared with the prior art, the present invention has the following beneficial effects:

[0042] The present invention divides the management area into regions and marks the risk level in each region, thereby generating a risk distribution map, so that relevant personnel in the region can timely understand the risk level in the current region and try to avoid the occurrence of infectious events.

[0043] The present invention adopts an optimized BP neural network to conduct risk assessment on each area. When the assessed risk of any area is higher than the initial risk level, a warning is issued in time, and then it is processed as soon as possible to reduce the probability of risk events. The BP neural network optimized by the particle swarm algorithm has the characteristics of high accuracy, fast calculation speed and good robustness, so that public health management has higher timeliness and more accurate assessment results, thereby maintaining a good public health status and reducing the risk of infection. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] Figure 1 is a step diagram of the public health management method in Example 1 of the present invention;

[0045] Figure 2 for Figure 1 Flowchart of optimization training of BP neural network model;

[0046] Figure 3 This is a schematic diagram of the structure of the public health management system of Example 2 of the present invention. DETAILED DESCRIPTION

[0047] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0048] It should be noted that when a component is referred to as being "mounted on" another component, it may be directly on the other component or there may be a central component. When a component is considered to be "set on" another component, it may be directly set on the other component or there may be a central component at the same time. When a component is considered to be "fixed to" another component, it may be directly fixed on the other component or there may be a central component at the same time.

[0049] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art to which the present invention belongs. The terms used herein in the specification of the present invention are only for the purpose of describing specific embodiments and are not intended to limit the present invention. The term "or / and" used herein includes any and all combinations of one or more of the related listed items.

[0050] Example 1: Please refer to Figure 1 This embodiment provides a public health management method, comprising the following steps:

[0051] S1: Collect geographic information of the management area, generate maps and divide the area.

[0052] In this embodiment, taking the layout of a hospital as an example, the hospital is divided into areas according to the floors, such as outpatient buildings, inpatient buildings, laboratory buildings, administrative buildings, etc., and the areas outside the floors are divided into corridor areas and other areas. For hospitals with more complex departments, a three-dimensional map can also be constructed to divide the floor areas of outpatient buildings, inpatient buildings, etc., and mark risks according to common conditions, such as marking floors with more infectious diseases as high-risk areas. In this embodiment, the risk levels are divided into 4 levels from low to high, including no risk area, low risk area, medium risk area, and high risk area.

[0053] S2: Collect big data based on personnel information, event information and environmental information, and conduct regional planning for personnel in the area.

[0054] Personnel information includes density data, health data, etc. Density data is the flow of people in each area during each time period. Health data includes the age of personnel, whether they carry infectious diseases, etc. Among them, personnel can be divided into patients, accompanying personnel and medical staff. In this embodiment, feasible areas are planned based on each person's health data. For example, patients without infectious diseases can stay in risk-free and low-risk areas, highly infectious patients and their accompanying personnel can only stay in high-risk areas, and medical staff can pass through the corresponding responsibility area, but when passing from high-risk areas to other areas, disinfection records should be uploaded.

[0055] Environmental information includes air data and ground data. Air data includes oxygen content, hazardous gas content, virus content, etc., among which oxygen content and hazardous gas content can be monitored by gas sensors. In order to reduce monitoring costs, disinfection frequency can be used instead of virus content. For example, for high-risk areas, comprehensive disinfection is carried out every half a day, and the disinfection data is recorded and uploaded. For low-risk areas, disinfection can be carried out every day or every few days, and the disinfection data is recorded and uploaded. Ground data mainly includes dust, ground garbage, stains, etc., and the sanitation status in each area can be collected in real time through monitoring. In this embodiment, the sanitation status is divided into multiple levels, and image recognition can be used to mark the sanitation level in each area.

[0056] Event information includes small event data based on individual infection and large event data based on collective infection. Binding event information with geographic information, environmental information, and personnel information facilitates analysis of the causes of events and helps predict the probability of the next event.

[0057] S3: Construct a risk assessment model based on BP neural network, input the current data of each area into the risk assessment model, and obtain the risk level in each area. BP neural network includes input layer, hidden layer and output layer.

[0058] First, the big data is normalized and the input of the BP neural network is obtained as follows:

[0059]

[0060] Where: X′ represents the result after data normalization, R max , R min represents the maximum and minimum values ​​of the normalized interval, X max , X min Represents the maximum and minimum values ​​of the input layer.

[0061] The normalized data is used as the input of the BP neural network, recorded as x(t), the output is set as y(t), and the sigmoid function is selected to construct the mapping relationship. The transfer function can be expressed as:

[0062]

[0063] The output of the hidden layer can be expressed as:

[0064]

[0065] Where: ij is the weight between the input layer and the hidden layer, θ j is the hidden layer threshold, and n is the number of hidden layer nodes.

[0066] The output of the output layer can be expressed as:

[0067]

[0068] Where: jk is the weight between the hidden layer and the output layer, θ k is the output layer threshold, and k is the number of output layer nodes.

[0069] The error between the actual output and the target output is defined as:

[0070]

[0071] Where: L k is the actual output, Z k is the target output, and m is the number of nodes in the input layer.

[0072] Apply the gradient descent method to update the weights, the expression is:

[0073]

[0074] In the formula, ω' ij is the updated weight between the input layer and the hidden layer, ω' jk is the updated weight between the hidden layer and the output layer, η is the learning efficiency, H j is the gradient.

[0075] Update weights, which can be expressed as:

[0076]

[0077] In the formula, θ' k is the updated output layer threshold, θ' j is the updated hidden layer threshold.

[0078] Particle swarm algorithm is introduced to optimize BP neural network.

[0079] If the particle is regarded as a solution vector in N-dimensional space, assuming that X and v represent the spatial position and velocity of particle C respectively, the particle update strategy can be expressed as:

[0080] v id (t+1)=ω d v id (t)+c 1 r 1 [P id (t)-X id (t)]+c 2 r 2 [P gd (t)-X id (t)]

[0081] X id (t+1)=X id (t)+v id (t)

[0082] Where: p id (t), p gd (t) are the individual optimal solution and the global optimal solution at time t, d is the dimension of the current particle, and d = 1, 2, ..., N, ω d is the particle’s inertia weight, c 1 、c 2 is the learning factor when the particle is updated, r 1 、r 2 is a random number uniformly distributed between [0, 1].

[0083] The traditional particle swarm algorithm is prone to fall into the local optimum and the phenomenon of "premature maturity" occurs, which affects the addressing accuracy. Therefore, in order to increase the cognitive ability and search range of particles, the PSO algorithm is improved. To this end, this embodiment designs an inertia weight adjustment method, namely:

[0084]

[0085] Where: L is the space length, X id is the coordinate value of the particle, is the average value of all particle coordinates. After improvement, during the addressing process, ω d It first gets bigger and then gets smaller, thus expanding the search range of the particles and solving the "premature maturity" problem.

[0086] Please combine Figure 2 , the optimized BP neural network is trained using the following method:

[0087] 1. Determine the fitness function of the particle swarm algorithm. The forward propagation theory of the BP neural network is used to calculate the fitness of the particles, and the fitness function of the particles is defined according to the mean square error obtained by training.

[0088] 2. Initialize the parameters of the BP neural network.

[0089] 3. Calculate the suitability and weight.

[0090] 4. Calculate particle velocity and position.

[0091] 5. Optimize the membership function.

[0092] 6. Update particle speed and position.

[0093] 7. Adjust the weights and thresholds of the BP neural network. The particle position vector will correspond to all the connection weights and thresholds of the BP algorithm, and find the optimal position through the fitness function, that is, find the optimal weights and thresholds of the BP network.

[0094] 8. Determine whether the test accuracy is met. If so, output the global optimal solution. Otherwise, repeat the above steps.

[0095] S4: Mark the map according to the risk level to obtain a risk distribution map.

[0096] In this embodiment, each risk level is marked with a different color. For example, the risk levels are marked as white, yellow, orange, and red from low to high, so that the risk level of each area can be quickly observed.

[0097] S5: Determine whether the risk level of each area exceeds the preset initial risk level, and issue an alarm if it does. Determine whether each person exceeds the passable area, and issue an alarm if it does.

[0098] Specifically, each area is initially marked with a level based on the patient group it receives, and the maximum risk level allowed for each area is recorded. If the risk assessment model determines that the risk level of the current area exceeds the maximum risk level allowed, an alarm is issued to relevant personnel, and emergency measures such as disinfection inspection and garbage cleaning are carried out in a timely manner.

[0099] When an alarm is issued for a regional risk level, analyze the cause of the alarm. This includes comparing the density of people and health data, as well as analyzing the environmental information in each corresponding area. For example, if an area has not been cleaned for a long time, resulting in excessive dust and garbage on the ground, the risk level will increase, and it is necessary to arrange for regional cleaning as soon as possible. Or if the population density in a certain area is too high and there are many infectious diseases, resulting in an increase in the risk level, then the relevant personnel can be arranged to evacuate, arrange personnel to enter the relevant area at different times, or focus on disinfecting the area.

[0100] When a person exceeds the feasible area, an early warning will be issued to the person. When the person exceeds the feasible area for a preset time, an early warning will be issued to the relevant staff. Of course, in actual management, accompanying persons are allowed to pass through the lower risk area, but they must not exceed the allowed stay time, otherwise they will also be warned.

[0101] The public health management method of this embodiment divides the management area into regions and marks the risk level in each region, thereby generating a risk distribution map, so that relevant personnel in the region can timely understand the risk level in the current region and try to avoid the occurrence of infectious events. At the same time, an optimized BP neural network is used to conduct risk assessment on each region. When the assessed risk of any region is higher than the initial risk level, an early warning is issued in time, and then it is processed as soon as possible to reduce the probability of risk events. The BP neural network optimized by the particle swarm algorithm has the characteristics of high accuracy, fast calculation speed and good robustness, so that public health management has higher timeliness and more accurate assessment results, thereby maintaining a good public health state and reducing the risk of infection.

[0102] Example 2: Please refer to Figure 3 This embodiment provides a public health management system that can be controlled using the public health management method of embodiment 1. The system includes a data acquisition module, a map generation module, a BP neural network model, a marking module, and an early warning module.

[0103] The data collection module is used to collect big data based on personnel information, event information, environmental information, and geographic information of the management area.

[0104] The map generation module is used to generate a management area map based on geographic information, and to divide the management area map into regions and mark initial risks. The regional division can be divided into regions based on the functions of each building area in the management area. The initial risk mark can also be marked according to the function of the building area, and the outpatient building, inpatient building, etc. can be further marked according to the department.

[0105] The BP neural network model is used to perform risk assessment on each sub-region divided in the management area according to the big data and the management area map, and output the risk level of each sub-region. The BP neural network includes an input layer, a hidden layer and an output layer. The big data is normalized and used as the input of the input layer. The BP neural network is optimized and trained using a particle swarm algorithm, and the optimization and training methods are as described in Example 1.

[0106] The marking module is used to mark the risk of the management area map according to the risk level. After the BP neural network model outputs the risk level of each sub-area, it is marked with corresponding colors to distinguish different risk areas. The marked risk distribution map can observe the risk status more intuitively.

[0107] The early warning module is used to determine whether the risk level of each area exceeds the preset initial risk level, and to issue an alarm if it does. It is also used to determine whether each person exceeds the passable area, and to issue an alarm if it does.

[0108] The technical features of the above-described embodiments may be arbitrarily combined. To make the description concise, not all possible combinations of the technical features in the above-described embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0109] The above-mentioned embodiments only express several implementation methods of the present invention, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the invention patent. It should be pointed out that, for ordinary technicians in this field, several variations and improvements can be made without departing from the concept of the present invention, and these all belong to the protection scope of the present invention. Therefore, the protection scope of the patent of the present invention shall be subject to the attached claims.

Claims

1. A public health management method, characterized in that: The control method comprises: S1: Collect geographic information of the management area, generate maps and divide the area; S2: Collect big data based on personnel information, event information and environmental information; conduct regional planning for personnel in the management area; S3: construct a risk assessment model based on BP neural network and use particle swarm algorithm to optimize the model; input the current data of each area into the risk assessment model to obtain the risk level in each area; S4: Marking the map according to the risk level to obtain a risk distribution map; S5: Determine whether the risk level of each area exceeds the preset initial risk level, and issue an alarm if so; determine whether each person exceeds the passable area, and issue an alarm if so.

2. A public health management method according to claim 1, characterized in that: In step S2, the personnel information includes density data and health data; the environmental information includes air data and ground data; and the event information includes small event data based on individual infection and large event data based on collective infection.

3. A public health management method according to claim 1, characterized in that: In step S2, the big data is normalized and expressed as: Where: X′ represents the result after data normalization, R max , R min represents the maximum and minimum values ​​of the normalized interval, X max , X min Represents the maximum and minimum values ​​of the input layer.

4. A public health management method according to claim 1, characterized in that: In step S3, the risk assessment model is constructed as follows: S31: construct an initial BP neural network; S32: Introduce particle swarm algorithm to optimize BP neural network; S33: Train the optimized BP neural network and retain parameters that meet the test accuracy.

5. A public health management method according to claim 4, characterized in that: In step S31, the BP network model includes an input layer, a hidden layer and an output layer; The output of the hidden layer is expressed as: Where: ij is the weight between the input layer and the hidden layer, θ j is the hidden layer threshold, n is the number of hidden layer nodes; The output expression of the output layer is: Where: jk is the weight between the hidden layer and the output layer, θ k is the output layer threshold, and k is the number of output layer nodes.

6. A public health management method according to claim 4, characterized in that: In step S32, the particle update strategy is expressed as: v id (t+1)=ω d v id (t)+c1r1[P id (t)-X id (t)]+c2r2[P gd (t)-X id (t)] X id (t+1)=X id (t)+v id (t) Where: p id (t), p gd (t) are the individual optimal solution and the global optimal solution at time t, d is the dimension of the current particle, and d = 1, 2, ..., N, ω d is the inertia weight of the particle, c1 and c2 are the learning factors when the particle is updated, and r1 and r2 are random numbers uniformly distributed between [0, 1]; The inertia weight of the particle swarm is adjusted, expressed as: Where: L is the space length, X id is the coordinate value of the particle, is the average value of all particle coordinates.

7. A public health management method according to claim 4, characterized in that: In step S32 and step S33, the training method of the BP neural network is as follows:

1. Determine the fitness function of the particle swarm algorithm; 2. Initializing parameters of the BP neural network; 3. Calculate the applicability and weight; 4. Calculate particle speed and position; 5. Optimize the membership function; 6. Update particle speed and position; 7. Adjust the weights and thresholds of the BP neural network; 8. Determine whether the test accuracy is met. If so, output the global optimal solution. Otherwise, repeat the above steps.

8. A public health management system, which adopts a public health management method as claimed in any one of claims 1 to 7, characterized in that: The control system comprises: Data collection module, used to collect big data based on personnel information, event information and environmental information, as well as geographic information of the management area; A map generation module, used to generate a management area map according to the geographic information, and to divide the management area map into regions and perform initial risk marking; A BP neural network model is used to perform risk assessment on each sub-region divided in the management area according to the big data and the management area map, and output the risk level of each sub-region; A marking module, used for marking the management area map with risk according to the risk level; The early warning module is used to determine whether the risk level of each area exceeds the preset initial risk level, and if so, an alarm is issued; and to determine whether each person exceeds the passable area, and if so, an alarm is issued.

Images