Real-time dynamic assessment method and device for hospital airborne infection risk, medium and product

Abstract

The invention discloses a hospital airborne infection risk real-time dynamic assessment method and device, a medium and a product, and relates to the technical field of public health, and the method comprises the steps: determining multi-person behavior parameters based on a person number index; performing feature extraction on the video monitoring data of the hospital by adopting a machine learning algorithm to generate a target detection image; obtaining a tracking target based on the personnel ID and the target bounding box in the target detection image by using a Byte Track algorithm; matching the tracking target with the target detection image; converting the multi-target tracking result into conversion data by using a perspective transformation method; an improved Wells-Riley model is adopted to determine the infection probability based on the multi-person behavior parameters and the conversion data; and performing real-time dynamic assessment on hospital airborne infection risks according to the infection probability. According to the method, the hospital airborne infection risk can be effectively assessed, and effective technical reference is provided for timely prevention and reduction of hospital infection possibility.

Description

Technical Field

[0001] The present application relates to the field of public health technology, and in particular to a real-time dynamic assessment method, equipment, medium and product for hospital airborne infection risk. Background Art

[0002] In recent years, the public has paid more and more attention to respiratory diseases. Respiratory diseases refer to diseases caused by pathogens invading the human body through the nose, throat, trachea and bronchi, etc. They are mainly spread by patients or infected people through droplets. Different respiratory diseases have different clinical manifestations. While treating patients, medical staff also face a huge risk of hospital-acquired infections. Hospital-acquired infections, also known as hospital-acquired infections, refer to non-primary infections acquired by people during their stay and treatment in the hospital.

[0003] As a place with a high density of people and frequent movement, hospitals have a high risk of nosocomial infection, especially when facing a complex and diverse group of people. Factors such as the number of people in the hospital space, contact duration, contact method, and contact distance will have a significant impact on the probability of infection. Although existing studies have provided a certain basis for the monitoring of nosocomial infection, there are still many problems that need to be solved: (1) Many studies focus on examining the risk of a single area, or only consider the impact of a single behavior or static state of a person in a single space on the assessment of infection risk. (2) Although some studies have introduced distance indexes to quantify the impact of social distance on exposure, it is actually still related to the density of people and fine-grained behavior. Previous studies have failed to analyze the infection risk of different locations in the space through specific spatial location information and partition information. (3) Some studies usually only assume that the risk of virus transmission will decrease as the distance between individuals increases, ignoring the characteristics of droplets and aerosols containing viruses spreading in the air.

[0004] Therefore, traditional management methods cannot effectively assess the risk of airborne infections in hospitals, and cannot provide effective technical references for timely prevention and reduction of the possibility of hospital infections. Summary of the invention

[0005] The purpose of this application is to provide a real-time dynamic assessment method, equipment, medium and product for hospital airborne infection risk, which can effectively assess the hospital airborne infection risk and provide an effective technical reference for timely prevention and reduction of the possibility of hospital infection.

[0006] To achieve the above objectives, this application provides the following solutions:

[0007] In the first aspect, the present application provides a real-time dynamic assessment method for hospital airborne infection risk, comprising:

[0008] Determine multi-personnel behavior parameters based on a personnel quantity index; the personnel quantity index is the total number of personnel in the monitoring area; the multi-personnel behavior parameters include contact time, contact distance, contact method and number of personnel;

[0009] Obtaining video surveillance data of the hospital, and using a machine learning algorithm to extract features from the video surveillance data to generate a target detection image; the target detection image includes a person ID and a target boundary box;

[0010] Using the Byte Track algorithm, a tracking target is obtained based on the person ID and the target bounding box;

[0011] Matching the tracking target with the target detection image to obtain a multi-target tracking result;

[0012] Using a perspective transformation method to perform coordinate transformation on the multi-target tracking results to obtain transformation data;

[0013] Using a modified Wells-Riley model, determining the infection probability based on the multi-personnel behavior parameters and the conversion data;

[0014] Based on the infection probability, a real-time dynamic assessment of the hospital airborne infection risk is conducted.

[0015] Optionally, the machine learning algorithm is a YOLOv8 algorithm.

[0016] Optionally, using a perspective transformation method to perform coordinate transformation on the multi-target tracking result to obtain transformation data includes:

[0017] Obtaining the spatiotemporal coordinate information of all personnel targets in the multi-target tracking result;

[0018] A transformation matrix is ​​determined using a perspective transformation formula, and the transformation matrix is ​​used to transform the spatiotemporal coordinate information of all personnel targets into spatial coordinates to obtain the transformation data.

[0019] Optionally, the space coordinates obtained by conversion are expressed as:

[0020]

[0021] Where (u, v) represents the Cartesian coordinates of each frame of the multi-target tracking result, (u, v, 1) represents the homogeneous coordinates, M represents the transformation matrix, and a ij (i,j=1,2,3) are the elements of the transformation matrix M, and (x,y,z) represent the spatial coordinates.

[0022] Optionally, the construction process of the improved Wells-Riley model includes:

[0023] The Wells-Riley model is obtained, and the correction coefficient and ventilation rate are added as new parameters to the Wells-Riley model to obtain the improved Wells-Riley model; the correction coefficient includes a contact distance correction coefficient, a contact mode correction coefficient and a mask wearing correction coefficient.

[0024] Optionally, the infection probability determination formula is expressed as:

[0025]

[0026] Where P is the infection probability; P d is the relationship between the number of droplets exposed in the air and the distance, and is the correction coefficient of the contact distance. d represents the distance; I is the number of infected people; q is the virus generation rate of the infected people; S is the contact correction coefficient of the person inhaling the virus, θ1 is the spatial position angle between the infected person and the people in the same monitoring area, d m is the diameter of the mouth when coughing; p is the breathing rate of the person; k a Correction factor when the infected person wears a mask, k b is the correction factor when a person wears a mask; t is the duration of contact between the infected person and the people in the same monitoring area; Q is the ventilation rate; E z is the air distribution efficiency.

[0027] Optionally, based on the infection probability, a real-time dynamic assessment of hospital airborne infection risk is performed, including:

[0028] Based on the infection probability, a high-risk infection area is determined, and real-time evaluation and dynamic monitoring of personnel in the high-risk infection area are performed; the high-risk infection area is a monitoring area where the infection probability is higher than a set threshold.

[0029] In a second aspect, the present application provides a computer device comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of any of the above-described methods for real-time dynamic assessment of hospital airborne infection risks.

[0030] In a third aspect, the present application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of any of the above-mentioned methods for real-time dynamic assessment of hospital airborne infection risks.

[0031] In a fourth aspect, the present application provides a computer program product, including a computer program, which, when executed by a processor, implements the steps of any of the above-mentioned methods for real-time dynamic assessment of hospital airborne infection risks.

[0032] According to the specific embodiments provided in this application, this application has the following technical effects:

[0033] The present application provides a real-time dynamic assessment method, equipment, medium and product for the risk of airborne infection in hospitals. By using a machine learning algorithm to extract features from the video surveillance data, it is possible to capture the behavior trajectory of personnel based on computer vision and provide fine-grained behavior information for short-distance contact. By using an improved Wells-Riley model, the infection probability is determined based on the multi-personnel behavior parameters and the conversion data, and based on the infection probability, the real-time dynamic assessment of the risk of airborne infection in hospitals can be effectively achieved, thereby providing an effective technical reference for timely prevention and reduction of the possibility of nosocomial infection, and effectively reducing the risk of infection. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0035] Figure 1 A flowchart of a real-time dynamic assessment method for hospital airborne infection risk provided by an embodiment of the present application;

[0036] Figure 2 A schematic diagram of contact distance provided in an embodiment of the present application;

[0037] Figure 3 A schematic diagram of a contact method provided in an embodiment of the present application;

[0038] Figure 4 A schematic diagram of contact time provided for an embodiment of the present application;

[0039] Figure 5 A schematic diagram of the number of personnel provided in an embodiment of the present application;

[0040] Figure 6 A schematic diagram of the structure of a computer device provided in one embodiment of the present application. DETAILED DESCRIPTION

[0041] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0042] In order to make the above-mentioned objects, features and advantages of the present application more obvious and easy to understand, the present application is further described in detail below in conjunction with the accompanying drawings and specific implementation methods.

[0043] In an exemplary embodiment, Figure 1 As shown, a real-time dynamic assessment method for hospital airborne infection risk is provided, which is executed by a computer device, specifically, it can be executed by a computer device such as a terminal or a server alone, or it can be executed by a terminal and a server together. In the embodiment of the present application, the method is applied to a server as an example for explanation, including the following steps 100 to 106. Among them:

[0044] Step 100: Determine multi-personnel behavior parameters based on the personnel quantity index. The personnel quantity index is the total number of personnel in the monitoring area. The multi-personnel behavior parameters include contact time, contact distance, contact method and number of personnel.

[0045] Among them, the fine-grained behaviors contained in the trajectory of personnel flow are analyzed, and the parameters of contact distance, contact method, contact time, and number of personnel are introduced, such as Figures 2 to 5 As shown. Figure 3 As shown, the contact method may also include but is not limited to face-to-face contact, back-to-back contact, and side-to-side contact. Figure 2 Part (a) is a schematic diagram of the spatial distribution of contact distance. Figure 2 (b) is a schematic diagram of the contact distance space coordinates.

[0046] Step 101: Obtain video surveillance data of the hospital, and use a machine learning algorithm to extract features from the video surveillance data to generate a target detection image. The target detection image contains a person ID and a target boundary box. The video surveillance data of the hospital can be obtained by using a camera.

[0047] Step 102: Using the Byte Track algorithm, obtain the tracking target based on the person ID and the target bounding box.

[0048] Step 103: Match the tracking target and the target detection image to obtain a multi-target tracking result. This step can actually achieve target tracking under multiple machine visions (such as cameras).

[0049] Step 104: Use the perspective transformation method to perform coordinate transformation on the multi-target tracking results to obtain transformation data.

[0050] Step 105: Using the improved Wells-Riley model, the infection probability is determined based on multiple personnel behavior parameters and conversion data.

[0051] Step 106: Based on the infection probability, a real-time dynamic assessment of the hospital airborne infection risk is performed.

[0052] In another exemplary embodiment of the present application, the implementation process of the above step 101 is described by taking the YOLOv8 algorithm as an example of the machine learning algorithm used. Based on the video surveillance data obtained in the hospital, the YOLOv8 algorithm is used to identify relevant personnel and perform feature extraction to generate personnel IDs and target bounding boxes. Specifically:

[0053] (1) Obtain video frames from the hospital’s surveillance data and input them into the YOLOv8 algorithm.

[0054] (2) Use the YOLOv8 algorithm for person target detection and feature extraction.

[0055] (3) Generate a unique ID and bounding box for the target.

[0056] In another exemplary embodiment of the present application, the implementation process of step 104 can be described as follows:

[0057] (1) Obtain the spatiotemporal coordinate information of all human targets in the multi-target tracking results and express it in three-dimensional coordinates (u, v, t).

[0058] (2) The perspective transformation formula is used to determine the transformation matrix, and the transformation matrix is ​​used to convert the spatiotemporal coordinate information of all personnel targets into spatial coordinates to obtain the transformation data. Among them, the pixel coordinates are converted into actual coordinates, and the transformation matrix M coordinates are solved using the perspective transformation formula. The coordinate perspective transformation formula is as follows:

[0059]

[0060] Where (u,v) represents the Cartesian coordinates of each frame of the multi-target tracking result. (u,v,1) represents the homogeneous coordinates. In order to unify the standard, the bottom midpoint of the target tracking box in the YOLOv8 algorithm is used as the coordinate point of the row of people to draw a bird's-eye view. M represents the transformation matrix. a ij (i,j=1,2,3) are the elements of the transformation matrix M. (x,y,z) represent the spatial coordinates, which are the homogeneous coordinates after transformation, that is, the coordinates in the bird's-eye view. Its Cartesian coordinates (x',y') satisfy the following equation:

[0061]

[0062] In another exemplary embodiment of the present application, in step 105, the construction process of the improved Wells-Riley model includes: adding four new parameters, namely, contact distance, contact method, correction coefficient of wearing a mask, and ventilation rate, to the traditional Wells-Riley model to obtain an improved Wells-Riley model.

[0063] Relevant calculation parameters are obtained from the video data to perform real-time calculation of the infection probability. The infection probability refers to the probability of infection of susceptible persons, which is expressed as:

[0064]

[0065] Where P is the infection probability. d is the relationship between the number of droplets exposed in the air and the distance, is the contact distance correction coefficient, d represents the distance. I represents the number of infected persons, which defaults to 1. q represents the virus generation rate of infected persons, which is 142. S represents the correction factor for the inhalation of viruses by personnel, i.e., the correction factor for the contact mode of the inhalation of viruses by susceptible persons, which is related to the angle and distance. θ1 is the spatial position angle between the infected person and the people in the same monitoring area. m is the diameter of the mouth during coughing, set to 0.02m. p is the breathing rate of the person, set to 0.96. k a Correction factor when the infected person wears a mask, k b is the correction factor when a person wears a mask, k a and k b are all 0.5. t is the duration of contact between the infected person and people in the same monitoring area. Q is the ventilation rate. E z is the air distribution efficiency.

[0066] In another exemplary embodiment of the present application, the implementation process of step 106 can be described as follows:

[0067] (1) Personnel behavior analysis:

[0068] According to the spatial-temporal coordinate information of personnel, the distribution of the number of personnel in each monitoring area is analyzed. According to the spatial-temporal trajectory information of high-risk infected personnel, the contact time, contact method and contact distance between high-risk personnel and others are extracted to analyze the personnel behavior.

[0069] (2) Implementation assessment of airborne infection risk:

[0070] According to the infection probability of susceptible persons in each monitoring area, high-risk infection areas and infected people are evaluated and monitored in real time. Among them, high-risk infection areas are determined based on the infection probability, and people in high-risk infection areas are evaluated and monitored in real time and dynamically. High-risk infection areas are monitoring areas where the infection probability is higher than the set threshold.

[0071] Based on the above description, compared with the existing methods, the main advantages of this application are:

[0072] 1. Infection risk analysis based on fine-grained behavioral information: This application provides fine-grained behavioral information for short-distance contacts based on computer vision-based personnel behavior trajectory capture, which goes beyond previous research methods that rely on shallow behavioral indicators. In addition, the camera-based personnel behavior trajectory capture is used to assess the risk of infection, combining the on-site personnel mask wearing and on-site ventilation environment measurement to measure the ventilation rate. On-site measurements of ventilation rates reflect the objective physical environment, while maintaining social distance and wearing masks are subjectively controlled. In this case, by real-time monitoring of personnel activities and gatherings, based on fine-grained personnel contact analysis, immediate measures can be taken to increase ventilation rates or broadcast alarms to strengthen mask wearing, which can effectively reduce the probability of infection.

[0073] 2. Risk assessment of multiple people in multiple spaces: The method proposed in this application can work collaboratively between multiple cameras (i.e., machine vision). When a dangerous person disappears from the view of one camera and enters another camera, the method provided in this application can still be tracked and identified, greatly reducing the workload of medical staff.

[0074] 3. Identification of infection risk in the spatiotemporal dimension: The advantage of the behavior trajectory capture method based on machine vision proposed in this application is that it can obtain contact information and calculate risks in the spatiotemporal dimension. The trajectory capture method proposed in this application can analyze the exposure time and its spatial distribution, as well as the changes in infection risk in the temporal dimension.

[0075] In an exemplary embodiment, a computer device is provided. The computer device may be a server or a terminal. The internal structure diagram thereof may be as follows: Figure 6As shown. The computer device includes a processor, a memory, an input / output interface (Input / Output, referred to as I / O) and a communication interface. Among them, the processor, the memory and the input / output interface are connected through a system bus, and the communication interface is connected to the system bus through the input / output interface. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store real-time dynamic assessment data of hospital airborne infection risks. The input / output interface of the computer device is used to exchange information between the processor and an external device. The communication interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, a real-time dynamic assessment method for hospital airborne infection risks is implemented.

[0076] Those skilled in the art will understand that Figure 6 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components. In an exemplary embodiment, a computer device is provided, including a memory and a processor, wherein a computer program is stored in the memory, and the processor implements the steps in the above-mentioned method embodiments when executing the computer program.

[0077] In an exemplary embodiment, a computer-readable storage medium is provided, storing a computer program, and when the computer program is executed by a processor, the steps in the above method embodiments are implemented.

[0078] In an exemplary embodiment, a computer program product is provided, including a computer program, and when the computer program is executed by a processor, the steps in the above method embodiments are implemented.

[0079] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with relevant regulations.

[0080] Those of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to the memory, database or other medium used in the embodiments provided in the present application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM may be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM).

[0081] The database involved in each embodiment provided in this application may include at least one of a relational database and a non-relational database. The non-relational database may include a distributed database based on blockchain, etc., but is not limited thereto. The processor involved in each embodiment provided in this application may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, etc., but is not limited thereto.

[0082] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above 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.

[0083] This article uses specific examples to illustrate the principles and implementation methods of this application. The description of the above embodiments is only used to help understand the method and core ideas of this application. At the same time, for those skilled in the art, according to the ideas of this application, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as limiting this application.

Claims

1. A real-time dynamic assessment method for hospital airborne infection risk, characterized in that: The real-time dynamic assessment method for hospital airborne infection risk includes: Determine multi-personnel behavior parameters based on a personnel quantity index; the personnel quantity index is the total number of personnel in the monitoring area; the multi-personnel behavior parameters include contact time, contact distance, contact method and number of personnel; Obtaining video surveillance data of the hospital, and using a machine learning algorithm to extract features from the video surveillance data to generate a target detection image; the target detection image includes a person ID and a target boundary box; Using the Byte Track algorithm, a tracking target is obtained based on the person ID and the target bounding box; Matching the tracking target with the target detection image to obtain a multi-target tracking result; Using a perspective transformation method to perform coordinate transformation on the multi-target tracking results to obtain transformation data; Using a modified Wells-Riley model, determining the infection probability based on the multi-personnel behavior parameters and the conversion data; Based on the infection probability, a real-time dynamic assessment of the hospital airborne infection risk is conducted.

2. The real-time dynamic assessment method for hospital airborne infection risk according to claim 1, characterized in that: The machine learning algorithm is the YOLOv8 algorithm.

3. The real-time dynamic assessment method for hospital airborne infection risk according to claim 1, characterized in that: The coordinate transformation of the multi-target tracking result is performed using a perspective transformation method to obtain transformation data, including: Obtaining the spatiotemporal coordinate information of all personnel targets in the multi-target tracking result; A transformation matrix is ​​determined using a perspective transformation formula, and the transformation matrix is ​​used to transform the spatiotemporal coordinate information of all personnel targets into spatial coordinates to obtain the transformation data.

4. The real-time dynamic assessment method for hospital airborne infection risk according to claim 3 is characterized in that: The space coordinates obtained by transformation are expressed as: Where (u, v) represents the Cartesian coordinates of each frame of the multi-target tracking result, (u, v, 1) represents the homogeneous coordinates, M represents the transformation matrix, and a ij (i,j=1,2,3) are the elements of the transformation matrix M, and (x,y,z) represent the spatial coordinates.

5. The real-time dynamic assessment method for hospital airborne infection risk according to claim 1, characterized in that: The construction process of the improved Wells-Riley model includes: The Wells-Riley model is obtained, and the correction coefficient and ventilation rate are added as new parameters to the Wells-Riley model to obtain the improved Wells-Riley model; the correction coefficient includes a contact distance correction coefficient, a contact mode correction coefficient and a mask wearing correction coefficient.

6. The real-time dynamic assessment method for hospital airborne infection risk according to claim 1, characterized in that: The formula for determining the infection probability is expressed as: Where P is the infection probability; P d is the relationship between the number of droplets exposed in the air and the distance, and is the correction coefficient of the contact distance. d represents the distance; I is the number of infected people; q is the virus generation rate of the infected people; S is the contact correction coefficient of the person inhaling the virus, θ1 is the spatial position angle between the infected person and the people in the same monitoring area, d m is the diameter of the mouth when coughing; p is the breathing rate of the person; k a Correction factor when the infected person wears a mask, k b is the correction factor when a person wears a mask; t is the duration of contact between the infected person and the people in the same monitoring area; Q is the ventilation rate; E z is the air distribution efficiency.

7. The real-time dynamic assessment method for hospital airborne infection risk according to claim 1, characterized in that: Based on the infection probability, a real-time dynamic assessment of hospital airborne infection risk is conducted, including: Based on the infection probability, a high-risk infection area is determined, and real-time evaluation and dynamic monitoring of personnel in the high-risk infection area are performed; the high-risk infection area is a monitoring area where the infection probability is higher than a set threshold.

8. A computer device comprising: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the real-time dynamic assessment method for hospital airborne infection risk according to any one of claims 1 to 7.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the real-time dynamic assessment method for hospital airborne infection risk described in any one of claims 1 to 7 is implemented.

10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the real-time dynamic assessment method for hospital airborne infection risk described in any one of claims 1 to 7 is implemented.

Images