Intelligent monitoring and early warning method and system for emergency visitor flow

Abstract

The invention provides an intelligent monitoring and early-warning method and system for the traffic of people in emergency treatment, and the method comprises the steps: obtaining the monitoring data of monitoring equipment in each region of emergency treatment, and extracting a plurality of frames of images from the monitoring data; performing target detection, feature extraction and image cutting processing on each frame of image through a computer vision technology to obtain a pedestrian feature image; identity recognition, identity classification and labeling processing are carried out based on the pedestrian feature images, and the pedestrian feature images of which the labeling types are in-service employees are rejected from the early warning personnel queue to obtain a standard data set; performing model training based on the standard data set to obtain a people flow prediction model; analyzing and predicting the human traffic of each emergency treatment area through a human traffic prediction model to obtain a prediction result; and generating early warning information of different grades according to the prediction result and the alarm threshold, and executing corresponding emergency response plans for the early warning information of different grades. According to the method, medical staff and resources are allocated in advance, and the emergency response capability of a hospital is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent monitoring and early warning of human traffic, and in particular to an intelligent monitoring and early warning method and system for emergency human traffic. Background Art

[0002] Emergency patient flow has obvious volatility and uncertainty, and its characteristics are affected by many factors, such as time, season, emergencies, disease epidemic trends, etc. When the number of patients received by the emergency department in a short period of time exceeds its carrying capacity, it not only affects the quality of medical services, but also causes multiple harms to patients and medical staff. As a scientific management method, emergency area patient flow monitoring and early warning help emergency departments cope with daily operational challenges, and also provide strong support in emergency situations, ensuring that emergency departments can respond quickly and effectively to various needs and emergencies.

[0003] However, in the existing technology, since the flow of people is affected by seasonal factors, the existing models have low flexibility in dealing with seasonality and trends, and there is no corresponding emergency plan; and due to the lack of visualization tools, the management end cannot monitor the flow of people in the emergency room in real time. Therefore, the emergency area patient flow monitoring method in the existing technology does not provide timely warning and emergency plan response. Summary of the invention

[0004] In view of this, the present invention provides an intelligent monitoring and early warning method and system for emergency patient flow to solve the above problems.

[0005] The present invention provides an intelligent monitoring and early warning method for emergency pedestrian flow, comprising: acquiring monitoring data of monitoring equipment in each area of ​​the emergency department, and extracting multiple frames of images therefrom; performing target detection, feature extraction and image cropping processing on each frame of the image through computer vision technology to obtain a pedestrian feature image; performing identity recognition, identity classification and labeling processing based on the pedestrian feature image, removing the pedestrian feature images with the labeling type of on-the-job employees from the warning personnel queue, and obtaining a standard data set; performing model training based on the standard data set to obtain a pedestrian flow prediction model; analyzing and predicting the pedestrian flow in each area of ​​the emergency department through the pedestrian flow prediction model to obtain a prediction result; generating different levels of warning information according to the prediction result and the alarm threshold, and executing corresponding emergency response plans for the different levels of warning information.

[0006] In another implementation of the present invention, the target detection and feature extraction processing are performed on each frame of image by computer vision technology to obtain the behavioral characteristics of pedestrians in each frame of image, including: identifying and locating the pedestrians contained in each frame of image by YOLOv5 deep learning model; performing feature extraction processing on the pedestrians according to the identification and positioning results to obtain the behavioral characteristics of the pedestrians in each frame of image; cropping a feature image containing the pedestrians from each frame of image according to the movement trajectory of the pedestrians determined by the behavioral characteristics; and standardizing the feature image containing the pedestrians to obtain a pedestrian feature image.

[0007] In another implementation of the present invention, the model training is performed based on the standard data set to obtain a pedestrian flow prediction model, including: dividing the standard data set into a training set, a validation set and a test set; converting the behavioral features extracted from the standard data set into time series data; taking the behavioral features at each time point as the observation value of the time series to construct a time series; and performing model training according to the time series corresponding to the training set to obtain a pedestrian flow prediction model.

[0008] In another implementation of the present invention, it also includes: using the test set to evaluate the model performance of the pedestrian flow prediction model, and the model performance includes accuracy, recall rate, F1 score, and average precision mean.

[0009] In another implementation of the present invention, it also includes: setting the alarm threshold based on key time points and high traffic periods in historical monitoring data.

[0010] In another implementation of the present invention, the alarm threshold is set based on the key time points and high flow periods in the historical monitoring data, including: calculating the flow mean according to the key time points and high flow periods in the historical monitoring data:

[0011]

[0012] Where N is the number of data points, x i is the flow rate of each data point;

[0013] The standard deviation is calculated based on the flow mean, and the volatility of the data is reflected by the degree of dispersion between the data points and the flow mean:

[0014]

[0015] According to the flow mean and the standard deviation, different levels of warning thresholds are set, which are divided into a mild warning threshold, a moderate warning threshold, and a severe warning threshold.

[0016] In another implementation of the present invention, the mild warning threshold is expressed as:

[0017] Threshold low = Mean + σ

[0018] The moderate warning threshold is expressed as:

[0019] Threshold medium =Mean+2σ

[0020] The critical warning threshold is expressed as:

[0021] Threshold high =Mean+3σ

[0022] Among them, Mean is the flow mean and σ is the standard deviation.

[0023] Another aspect of the present invention provides an intelligent monitoring and early warning system for emergency pedestrian flow, including: a data acquisition module: acquiring monitoring data from monitoring equipment in each emergency area, and extracting multiple frames of images therefrom; a data processing module: performing target detection, feature extraction and image cropping processing on each frame of the image through computer vision technology to obtain a pedestrian feature image; performing identity recognition, identity classification and labeling processing based on the pedestrian feature image, removing the pedestrian feature images with the label type of on-the-job employees from the warning personnel queue, and obtaining a standard data set; a model training module: performing model training based on the standard data set to obtain a pedestrian flow prediction model; a pedestrian flow prediction module: analyzing and predicting the pedestrian flow in each emergency area through the pedestrian flow prediction model to obtain a prediction result; an early warning module: generating early warning information of different levels according to the prediction result and the alarm threshold, and executing corresponding emergency response plans for the early warning information of different levels.

[0024] Another aspect of the present invention provides an electronic device, comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the steps of an intelligent monitoring and early warning method for emergency patient flow are implemented as described in any one of the above items.

[0025] Another aspect of the present invention provides a computer storage medium, characterized in that a computer program is stored on the computer storage medium, and when the computer program is executed by a processor, the steps of the intelligent monitoring and early warning method for emergency patient flow as described in any one of the above items are implemented.

[0026] The intelligent monitoring and early warning method for emergency patient flow of the present invention solves the problems of untimely early warning of increased emergency patient flow and untimely response to emergency plans. It can timely monitor the increase in emergency patient flow, thereby providing early warning, improving the efficiency of emergency management and medical services, and realizing early deployment of medical personnel and resources through patient flow early warning, thereby improving the hospital's emergency response capability. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following briefly introduces the drawings required for use in the embodiments or the prior art description. By reading the detailed description of the following implementation, the advantages and benefits of the solutions become clear to those skilled in the art. The drawings are only used to illustrate the preferred implementation and are not considered to be limitations of the present invention. In the drawings:

[0028] Figure 1 The present invention is a flowchart of an intelligent monitoring and early warning method for emergency patient flow according to an embodiment of the present invention.

[0029] Figure 2 The figure is a schematic diagram of a data processing flow of an embodiment of the present invention.

[0030] Figure 3 The figure is a schematic diagram of the model training process of an embodiment of the present invention.

[0031] Figure 4 The figure is a schematic diagram of a warning level setting process according to an embodiment of the present invention. DETAILED DESCRIPTION

[0032] In order to enable those skilled in the art to better understand the technical solutions in the embodiments of the present invention, the technical solutions in the embodiments of the present invention will be described clearly and in detail below in conjunction with the drawings 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. All other embodiments obtained by ordinary technicians in the field based on the embodiments in the embodiments of the present invention should fall within the scope of protection of the embodiments of the present invention.

[0033] Figure 1 A flow chart of an intelligent monitoring and early warning method for emergency patient flow provided by an embodiment of the present invention is shown in FIG. Figure 1 As shown, this embodiment mainly includes:

[0034] S101, acquiring monitoring data from monitoring equipment in each emergency area, and extracting multiple frames of images therefrom.

[0035] For example, the emergency area is subdivided into multiple sub-areas, including the triage area, waiting area, and treatment area, and their flow and usage are monitored separately. The monitoring equipment in each area of ​​the emergency department is labeled with camera parameters, and the monitoring data of each area of ​​the emergency department is collected in real time and labeled with parameters, and multiple frames of image data are extracted from them.

[0036] S102, performing target detection, feature extraction and image cropping processing on each frame of the image through computer vision technology to obtain a pedestrian feature image.

[0037] S103, performing identity recognition, identity classification and labeling processing based on the pedestrian feature images, removing the pedestrian feature images with the label type of on-the-job employees from the warning personnel queue, and obtaining a standard data set.

[0038] For example, the facial features or body features of pedestrians are identified in the pedestrian feature image, and the pedestrian identity is classified and labeled in combination with the hospital staff information database to distinguish between patients and hospital employees, and hospital employees are excluded from the early warning queue to improve the accuracy of early warning. The standard data set contains annotated pedestrian feature images and corresponding labels, and the pedestrian feature images include images in different environments and different lighting conditions.

[0039] S104: Perform model training based on the standard data set to obtain a pedestrian flow prediction model.

[0040] S105. Analyze and predict the flow of people in each emergency area using the flow of people prediction model to obtain a prediction result.

[0041] Exemplarily, the crowd flow prediction model is used to analyze and predict the crowd flow in each emergency area, determine the crowd flow counting area of ​​each emergency area, and count the crowd flow monitoring data of several areas based on the movement trajectory data of pedestrians, wherein the crowd flow monitoring data includes the number of pedestrians in the crowd flow counting area and the number of pedestrians leaving the counting area.

[0042] S106: Generate warning information of different levels according to the prediction results and the alarm threshold, and execute corresponding emergency response plans for the warning information of different levels.

[0043] For example, based on the real-time flow of people in the emergency room, multi-level warnings are generated through warning thresholds, and corresponding emergency response plans are generated according to different warning levels. Managers can view the flow of people data, historical trends and warning information in real time through mobile terminals, thereby realizing the early deployment of medical staff and resources and improving the hospital's emergency response capabilities.

[0044] The contents of the emergency response plan mainly include human resource allocation, personnel diversion management, logistics support, and external support coordination.

[0045] The human resource deployment is based on the system detecting that the flow of people has reached a threshold of a corresponding level, and the system notifies the corresponding personnel of the emergency team by calling the user through the mobile terminal.

[0046] The personnel triage management dynamically adjusts the triage standards according to the urgency of the person, ensuring that critically ill patients receive priority treatment, and reminds department heads through mobile terminals to quickly see patients by immediately adding consulting rooms, etc.

[0047] The logistics support is based on the system detecting that the flow of people has reached a threshold of a corresponding level. The system notifies the logistics and cleaning department by calling the user through the mobile terminal to strengthen the cleaning force and avoid cross infection.

[0048] The external support coordination, based on the system detecting that the flow of people has reached a threshold of a corresponding level, reminds the department head through a mobile terminal whether to request support from other health departments.

[0049] The intelligent monitoring and early warning method for emergency patient flow of the present invention solves the problems of untimely early warning of increased emergency patient flow and untimely response to emergency plans. It can timely monitor the increase in emergency patient flow, thereby providing early warning, improving the efficiency of emergency management and medical services, and realizing early deployment of medical personnel and resources through patient flow early warning, thereby improving the hospital's emergency response capability.

[0050] In another implementation of the present invention, Figure 2 As shown, the target detection and feature extraction processing are performed on each frame of the image by computer vision technology to obtain the behavior characteristics of pedestrians in each frame of the image, including:

[0051] S201. Identify and locate pedestrians in each frame of the image using the YOLOv5 deep learning model.

[0052] S202: Perform feature extraction on the pedestrian according to the recognition and positioning results to obtain the behavior features of the pedestrian in each frame of the image.

[0053] S203 , cutting out a feature image containing the pedestrian from each frame of image according to the moving trajectory of the pedestrian determined by the behavior characteristics.

[0054] S204 performs standardization processing on the feature image containing pedestrians to obtain a pedestrian feature image.

[0055] In another implementation of the present invention, Figure 3 As shown, the model training is performed based on the standard data set to obtain a pedestrian flow prediction model, including:

[0056] S301, dividing the standard data set into a training set, a validation set and a test set.

[0057] S302: Convert the behavior features extracted from the standard data set into time series data.

[0058] S303: Taking the behavior characteristics of each time point as the observation value of the time series, constructing the time series.

[0059] S304: Perform model training according to the time series corresponding to the training set to obtain a pedestrian flow prediction model.

[0060] Exemplarily, data preprocessing is performed to ensure the accuracy of analysis results by cleaning outliers and missing data in the data; data quality and the generalization ability of the model are improved through cleaning, standardization and enhancement processing; and the obtained data set is cleaned.

[0061] Use the tool library to divide the data into training set, validation set and test set to evaluate the model performance. Usually, the data is divided in the proportion of 80% training set, 10% validation set and 10% test set. After the model is trained and evaluated, it is deployed in the actual environment, using the cloud server as the deployment platform, loaded and used for real-time video stream processing, and the processing results are displayed to users or saved to the database for recording.

[0062] In another implementation of the present invention, it also includes: using the test set to evaluate the model performance of the pedestrian flow prediction model, and the model performance includes accuracy, recall rate, F1 score, and average precision mean.

[0063] Exemplarily, accuracy refers to the proportion of correctly identified objects, recall refers to the proportion of correctly identified people, F1 score refers to an indicator that comprehensively considers accuracy and recall, and mean average precision refers to a standard evaluation indicator in object detection.

[0064] In another implementation of the present invention, it also includes: setting the alarm threshold based on key time points and high traffic periods in historical monitoring data.

[0065] In another implementation of the present invention, Figure 4 As shown, the alarm threshold is set based on the key time points and high traffic periods in the historical monitoring data, including:

[0066] S401. Calculate the traffic mean based on the key time points and high traffic periods in the historical monitoring data:

[0067]

[0068] Where N is the number of data points, x i is the flow rate of each data point.

[0069] S402, calculating the standard deviation value according to the flow mean, and reflecting the volatility of the data through the degree of dispersion of the data points and the flow mean:

[0070]

[0071] S403. According to the flow mean and the standard deviation, different levels of warning thresholds are set, which are divided into a mild warning threshold, a moderate warning threshold, and a severe warning threshold.

[0072] In another implementation of the present invention, the mild warning threshold is expressed as:

[0073] Threshold low = Mean + σ

[0074] The moderate warning threshold is expressed as:

[0075] Threshold medium =Mean+2σ

[0076] The critical warning threshold is expressed as:

[0077] Threshold high =Mean+3σ

[0078] Among them, Mean is the flow mean and σ is the standard deviation.

[0079] In another implementation of the present invention, the prediction results are displayed on a mobile terminal, and the manager can view the human flow data, future trends and warning information in real time through the mobile terminal, so that the manager can understand the flow of emergency patients in real time.

[0080] The mobile display can show the flow of people in a certain area. Patients and hospital employees are displayed in different colors. Different colors represent different groups of people. For example, white can represent doctors, blue represents nurses, purple represents other staff, red represents patients, etc.

[0081] Through real-time monitoring, predictive analysis, optimized resource allocation, patient experience and safety management and other means, the efficiency and quality of emergency services can be significantly improved.

[0082] Another aspect of the present invention provides an intelligent monitoring and early warning system for emergency patient flow, comprising:

[0083] Data acquisition module: obtains monitoring data from monitoring equipment in each emergency area and extracts multiple frames of images from it.

[0084] Data processing module: Use computer vision technology to perform target detection, feature extraction and image cropping on each frame of the image to obtain a pedestrian feature image; perform identity recognition, identity classification and labeling based on the pedestrian feature image, remove the pedestrian feature images with the label type of on-the-job employees from the warning personnel queue, and obtain a standard data set.

[0085] Model training module: Perform model training based on the standard data set to obtain a pedestrian flow prediction model.

[0086] Crowd flow prediction module: The crowd flow prediction model is used to analyze and predict the crowd flow in each emergency area to obtain prediction results.

[0087] Early warning module: generates early warning information of different levels according to the prediction results and alarm thresholds, and executes corresponding emergency response plans for the early warning information of different levels.

[0088] In another aspect of the present invention, an electronic device includes: a processor, a memory, a communication bus, and a communication interface.

[0089] in:

[0090] The processor, memory and communication interface communicate with each other through a communication bus.

[0091] Communication interface, used to communicate with other electronic devices or servers.

[0092] The processor is used to execute the program, and specifically can execute the steps of any one of the emergency patient flow intelligent monitoring and early warning methods in the above-mentioned embodiments.

[0093] Specifically, the program may include program codes including computer operation instructions.

[0094] The processor may be a central processing unit (CPU), or an application specific integrated circuit (ASIC), or one or more integrated circuits configured to implement the embodiments of the present application. The one or more processors included in the smart device may be processors of the same type, such as one or more CPUs; or processors of different types, such as one or more CPUs and one or more ASICs.

[0095] The memory is used to store programs. The memory may include a high-speed RAM memory, and may also include a non-volatile memory, such as at least one disk memory.

[0096] The program can be specifically used to enable the processor to execute to implement the steps of any one of the emergency flow intelligent monitoring and early warning methods described in the embodiments. The specific implementation of each step in the program can refer to the corresponding descriptions in the steps and units executed by any one of the emergency flow intelligent monitoring and early warning methods in the above steps, which will not be repeated here. Those skilled in the art can clearly understand that for the convenience and simplicity of description, the specific working process of the above-described equipment and modules can refer to the corresponding process description in the aforementioned method embodiment.

[0097] The method according to the embodiment of the present invention may be implemented in a server equipped with a central processing unit (CPU) and an image processing unit (COU).

[0098] Thus far, specific embodiments of the present invention have been described. Other embodiments are within the scope of the appended claims. In some cases, the actions recited in the claims can be performed in a different order and still achieve the desired results. In addition, the processes depicted in the accompanying drawings do not necessarily require the specific order or sequential order shown to achieve the desired results.

[0099] It should be noted that all directional indications in the embodiments of the present invention (such as up, down, left, right, back, etc.) are only used to explain the relative position relationship between the components in a certain specific order (as shown in the accompanying drawings). If the specific order changes, the directional indication will also change accordingly.

[0100] In the description of the present invention, the terms "first" and "second" are only used to facilitate the description of different components or names, and cannot be understood as indicating or implying a sequential relationship, relative importance, or implicitly indicating the number of the indicated technical features. Therefore, the features defined as "first" and "second" may explicitly or implicitly include at least one of the features.

[0101] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art of the present invention. The terms used in the specification of the present invention herein are only for the purpose of describing specific embodiments and are not intended to limit the present invention.

[0102] It should be noted that although the specific embodiments of the present invention are described in detail in conjunction with the accompanying drawings, it should not be understood as limiting the scope of protection of the present invention. Within the scope described in the claims, various modifications and variations that can be made by those skilled in the art without creative work still belong to the scope of protection of the present invention.

[0103] The examples of the embodiments of the present invention are intended to concisely illustrate the technical features of the embodiments of the present invention so that those skilled in the art can intuitively understand the technical features of the embodiments of the present invention, and are not intended to be improper limitations of the embodiments of the present invention.

[0104] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. An intelligent monitoring and early warning method for emergency patient flow, characterized in that: include: Obtain monitoring data from monitoring equipment in each area of ​​the emergency department and extract multiple frames of images from them; Computer vision technology is used to perform target detection, feature extraction and image cropping on each frame of the image to obtain a pedestrian feature image; Based on the pedestrian feature images, identity recognition, identity classification and labeling are performed, and pedestrian feature images with the label type of on-the-job employees are removed from the warning personnel queue to obtain a standard data set; Performing model training based on the standard data set to obtain a pedestrian flow prediction model; Analyze and predict the flow of people in each emergency area through the flow prediction model to obtain prediction results; Different levels of warning information are generated according to the prediction results and the alarm threshold, and corresponding emergency response plans are executed for the different levels of warning information.

2. The method according to claim 1, characterized in that The target detection and feature extraction processing are performed on each frame of the image by computer vision technology to obtain the behavior characteristics of pedestrians in each frame of the image, including: The pedestrians in each frame are identified and located using the YOLOv5 deep learning model. Perform feature extraction on pedestrians based on the recognition and positioning results to obtain the behavior features of pedestrians in each frame of image; According to the moving trajectory of the pedestrian determined by the behavior characteristics, a feature image containing the pedestrian is cut out from each frame of the image; The pedestrian feature image is standardized to obtain a pedestrian feature image.

3. The method according to claim 1, characterized in that The model training is performed based on the standard data set to obtain a pedestrian flow prediction model, including: Dividing the standard data set into a training set, a validation set and a test set; Converting the behavioral features extracted from the standard data set into time series data; The behavioral characteristics of each time point are used as the observation value of the time series to construct the time series; Model training is performed according to the time series corresponding to the training set to obtain a pedestrian flow prediction model.

4. The method according to claim 3, characterized in that Also includes: The test set is used to evaluate the model performance of the pedestrian flow prediction model, wherein the model performance includes accuracy, recall, F1 score, and average precision mean.

5. The method according to claim 1, characterized in that Also includes: The alarm threshold is set based on key time points and high traffic periods in historical monitoring data.

6. The method according to claim 5, characterized in that The step of setting the alarm threshold based on the key time points and high traffic periods in the historical monitoring data includes: Calculate the traffic mean based on key time points and high traffic periods in historical monitoring data: Where N is the number of data points, x i is the flow rate of each data point; The standard deviation is calculated based on the flow mean, and the volatility of the data is reflected by the degree of dispersion between the data points and the flow mean: According to the flow mean and the standard deviation, different levels of warning thresholds are set, which are divided into a mild warning threshold, a moderate warning threshold, and a severe warning threshold.

7. The method according to claim 6, characterized in that The mild warning threshold is expressed as: Threshold low =Mean+σ The moderate warning threshold is expressed as: Threshold medium =Mean+2σ The critical warning threshold is expressed as: Threshold high =Mean+3σ Among them, Mean is the flow mean and σ is the standard deviation.

8. An intelligent monitoring and early warning system for emergency patient flow, characterized in that: include: Data acquisition module: obtains monitoring data from monitoring equipment in each area of ​​the emergency department and extracts multiple frames of images from them; Data processing module: Use computer vision technology to perform target detection, feature extraction and image cropping on each frame of image to obtain pedestrian feature images; perform identity recognition, identity classification and labeling based on the pedestrian feature images, remove pedestrian feature images with the label type of on-the-job employees from the warning personnel queue, and obtain a standard data set; Model training module: performing model training based on the standard data set to obtain a crowd flow prediction model; Crowd flow prediction module: Analyze and predict the flow of people in each emergency area through the crowd flow prediction model to obtain prediction results; Early warning module: generates early warning information of different levels according to the prediction results and alarm thresholds, and executes corresponding emergency response plans for the early warning information of different levels.

9. An electronic device, characterized in that: include: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the steps of an intelligent monitoring and early warning method for emergency patient flow are implemented as described in any one of claims 1 to 7.

10. A computer storage medium, characterized in that: The computer storage medium stores a computer program, and when the computer program is executed by the processor, the steps of the intelligent monitoring and early warning method for emergency patient flow as described in any one of claims 1 to 7 are implemented.

Images