Medical waste tracking management system and method combined with face recognition

Abstract

The invention discloses a medical waste tracking management system and method combined with face recognition, and relates to the technical field of medical waste management, and the system comprises the steps: obtaining target medical waste; calling a predetermined feature index to perform multi-source feature collection to obtain target feature index information, and generating an initial point code label; similar clustering of the target medical waste is carried out; matching and determining a predetermined waste treatment decision according to the waste clustering result; calling an intelligent image collector to carry out dynamic monitoring to obtain a target processing image; matching the feature information of the target personnel in a personnel database; and obtaining a target point code label. The technical problems that in the prior art, medical waste information tracking is not accurate and complete enough, dynamic monitoring of the treatment process is lacked, and then the traceability, accuracy and safety of the medical waste treatment process are insufficient are solved, and the technical effect of improving the traceability, accuracy and safety of the waste treatment process is achieved.

Description

Technical Field

[0001] The present application relates to the technical field related to medical waste management, and specifically to a medical waste tracking management system and method combined with face recognition. Background Art

[0002] With the increasingly stringent requirements for medical waste treatment and the complexity of management, existing medical waste management methods face many challenges, especially in the tracking and monitoring links. Medical waste usually includes needles, medicine bottles, surgical instruments, etc. Its management involves multiple levels such as environmental protection and public health safety. In recent years, the medical industry has gradually introduced Internet of Things technology, barcode scanning technology, smart sensors, etc., and carried out real-time tracking and management of medical waste through equipment automation monitoring and data collection. However, there are still some limitations. For example, the treatment of the various and complex types of medical waste is not detailed and complete enough, and it is impossible to effectively realize the real-time monitoring and tracking of personnel behavior, which in turn affects the transparency of medical waste treatment, the ability to trace responsibility, and the intelligent management of the waste treatment process. Improper treatment may also lead to serious health and safety risks.

[0003] Therefore, at the current stage, the technologies related to medical waste tracking and management have technical problems such as inaccurate and complete tracking of medical waste information and lack of dynamic monitoring of the treatment process, which in turn leads to insufficient traceability, accuracy and safety of the medical waste treatment process. Summary of the invention

[0004] This application solves the technical problems in the prior art of inaccurate and complete tracking of medical waste information and lack of dynamic monitoring of the treatment process, which in turn leads to insufficient traceability, accuracy and safety of the medical waste treatment process by providing a medical waste tracking management system and method combined with facial recognition, thereby achieving the technical effect of improving the traceability, accuracy and safety of the waste treatment process.

[0005] The present application provides a medical waste tracking and management system combined with face recognition, the system comprising: a target medical waste acquisition module, used to acquire target medical waste, the target medical waste refers to any waste that meets the predetermined waste constraints; a target feature indicator information collection module, used to retrieve the predetermined feature indicators to perform multi-source feature collection on the target medical waste to obtain target feature indicator information, and generate an initial point code label for the target feature indicator information; a waste clustering result acquisition module, used to perform similar clustering of the target medical waste with the target feature indicator information as a constraint to obtain a waste clustering result; a predetermined waste treatment decision determination module, used to match the target medical waste according to the waste clustering result; Determine the predetermined waste disposal decision of the target medical waste; a target processing image acquisition module, used to call the intelligent image collector to dynamically monitor the treatment process of the target medical waste based on the predetermined waste disposal decision, and obtain a target processing image; a target person feature information matching module, used to extract the target face slice in the target processing image based on face recognition technology, and match the target person feature information of the target person corresponding to the target face slice in the personnel database; a target point code label acquisition module, used to add the target person feature information to the initial point code label to obtain a target point code label, wherein the target point code label is used to track and manage the target medical waste.

[0006] In a possible implementation, the medical waste tracking and management method combined with facial recognition also performs the following processing: the predetermined characteristic indicators include at least the source, transmission route, pathogen type and risk level.

[0007] In a possible implementation, the medical waste tracking and management method combined with facial recognition also performs the following processing: collecting a target image of the target medical waste; identifying text on the target image to obtain target label text information; analyzing the target label text information to determine the target source of the target medical waste; and adding the target source to the target characteristic indicator information.

[0008] In a possible implementation, the medical waste tracking and management method combined with face recognition also performs the following processing: obtaining first characteristic indicator information of first medical waste; when the target characteristic indicator information and the first characteristic indicator information meet the predetermined clustering constraints, forming a first waste clustering cluster based on the target medical waste and the first medical waste; and forming the waste clustering result according to the first waste clustering cluster.

[0009] In a possible implementation, a predetermined weight distribution of the predetermined characteristic indicator is obtained; the target characteristic indicator information and the first characteristic indicator information are analyzed and calculated in sequence in combination with the predetermined weight distribution to obtain a target characteristic coefficient of the target medical waste and a first characteristic coefficient of the first medical waste respectively; the target characteristic coefficient is compared with the first characteristic coefficient to obtain a first characteristic deviation rate; a cluster support evaluation function is introduced to analyze the first characteristic deviation rate to obtain a first cluster support; it is determined whether the first cluster support meets a predetermined support threshold; if not, the first waste cluster is adjusted and intervened.

[0010] In a possible implementation, the medical waste tracking and management method combined with face recognition further performs the following processing: the expression of the cluster support evaluation function is as follows: ; in, refers to the first characteristic coefficient The target characteristic coefficient The first cluster support, k refers to the magnitude of the cluster support evaluation function, and , is the clustering decay factor, and .

[0011] In a possible implementation, the medical waste tracking and management method combined with face recognition also performs the following processing: extracting the first image from the multiple images, the first image corresponding to the first processing stage; obtaining the first face slice in the first image using the semantic recognition principle; matching the first person feature information of the first person corresponding to the first face slice in the personnel database; forming the target person based on the first person, and forming the target person feature information based on the first person feature information.

[0012] The present application also provides a medical waste tracking and management method combined with face recognition, the method comprising: obtaining target medical waste, the target medical waste refers to any waste that meets the predetermined waste constraint; calling predetermined feature indicators to perform multi-source feature collection on the target medical waste to obtain target feature indicator information, and generating an initial point code label of the target feature indicator information; performing similarity clustering of the target medical waste with the target feature indicator information as a constraint to obtain a waste clustering result; determining a predetermined waste treatment decision for the target medical waste based on the waste clustering result matching; calling an intelligent image collector to dynamically monitor the treatment process of the target medical waste based on the predetermined waste treatment decision to obtain a target treatment image; extracting a target face slice in the target treatment image based on face recognition technology, and matching the target personnel feature information of the target personnel corresponding to the target face slice in a personnel database; adding the target personnel feature information to the initial point code label to obtain a target point code label, wherein the target point code label is used to track and manage the target medical waste.

[0013] Through the medical waste tracking management system and method combined with face recognition proposed in this application, it is intended to obtain the target medical waste; retrieve the predetermined characteristic indicators for multi-source characteristic collection to obtain the target characteristic indicator information, and generate the initial point code label; perform similar clustering of the target medical waste; determine the predetermined waste treatment decision based on the waste clustering result matching; call the intelligent image collector for dynamic monitoring to obtain the target treatment image; match the target personnel characteristic information in the personnel database; and obtain the target point code label. It solves the technical problems existing in the prior art that the medical waste information tracking is not accurate and complete, and there is a lack of dynamic monitoring of the treatment process, which leads to insufficient traceability, accuracy and safety of the medical waste treatment process, and achieves the technical effect of improving the traceability, accuracy and safety of the waste treatment process. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] In order to more clearly illustrate the technical solution of the embodiment of the present disclosure, the accompanying drawings of the embodiment of the present disclosure will be briefly introduced below. A flow chart is used in the present application to illustrate the operations performed by the system according to the embodiment of the present application. It should be understood that the preceding or following operations are not necessarily performed accurately in order. On the contrary, various steps can be processed in reverse order or simultaneously as needed. At the same time, other operations can also be added to these processes, or one or more operations can be removed from these processes.

[0015] Figure 1 A schematic diagram of the structure of a medical waste tracking and management system combined with face recognition provided in an embodiment of the present application; Figure 2A flowchart of a medical waste tracking and management method combined with face recognition provided in an embodiment of the present application.

[0016] Explanation of the reference numerals: target medical waste acquisition module 10, target characteristic index information collection module 20, waste clustering result acquisition module 30, predetermined waste treatment decision determination module 40, target treatment image acquisition module 50, target personnel characteristic information matching module 60, target point code label acquisition module 70. DETAILED DESCRIPTION

[0017] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below.

[0018] In order to make the objectives, technical solutions and advantages of the present application clearer, the present application will be further described in detail below in conjunction with the accompanying drawings. The described embodiments should not be regarded as limiting the present application. All other embodiments obtained by ordinary technicians in the field without making creative work are within the scope of protection of this application.

[0019] In the following description, reference is made to "some embodiments", which describe a subset of all possible embodiments, but it is understood that "some embodiments" may be the same subset or different subsets of all possible embodiments, and may be combined with each other without conflict, and the terms "first\second" involved are merely to distinguish similar objects and do not represent a specific ordering of objects. The terms "including" and "having" and any variations are intended to cover non-exclusive inclusions, for example, a process, system, product, or server that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or modules that are not clearly listed or inherent to these processes, products, or devices. Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those generally understood by technicians in the technical field of this application. The terms used herein are for the purpose of describing the embodiments of the present application only.

[0020] The present application embodiment provides a medical waste tracking and management system combined with face recognition, such as Figure 1 As shown, the system comprises: The target medical waste acquisition module 10 is used to acquire target medical waste, where the target medical waste refers to any waste that meets a predetermined waste constraint.

[0021] Preferably, target medical waste is obtained, and target medical waste refers to any waste that meets the predetermined waste constraints among multiple medical wastes. For example, medical waste can be divided into different types according to its properties, such as infectious waste (such as needles and medical devices that have been in contact with infectious disease patients), harmless waste, pharmaceutical waste, chemical waste, plastic bags and ordinary bedding, etc. The target medical waste meets the standards of the predetermined waste constraints and can be managed more accurately through identification and tracking.

[0022] The target characteristic indicator information collection module 20 is used to retrieve predetermined characteristic indicators to perform multi-source characteristic collection on the target medical waste to obtain target characteristic indicator information, and generate an initial point code label of the target characteristic indicator information.

[0023] The specific configuration of the target characteristic indicator information collection module 20 further includes that the predetermined characteristic indicators at least include the source, the route of infection, the type of pathogen and the risk level.

[0024] Preferably, comprehensive characteristic information is collected by retrieving multiple key characteristic indicators of the target medical waste (such as source, transmission route, pathogen type, risk level, etc.), and an initial point code label (such as a QR code or barcode) is generated based on this information. The label is used to uniquely identify the target medical waste. Specifically, the predetermined characteristic indicators refer to the key information that needs to be collected in order to accurately describe and track the target waste during the medical waste management process. These indicators can help distinguish and identify important characteristics of the waste, such as type, source, infectiousness, risk level, etc., and thus provide a basis for subsequent management and treatment. The predetermined characteristic indicators at least include source, transmission route, pathogen type and risk level, among which source refers to the source type of medical waste. Type, which may be from different medical institutions such as hospitals, clinics, laboratories, for example, whether the waste is generated by departments such as infectious diseases and emergency departments; transmission route refers to the possible transmission mode of waste, usually including air transmission, contact transmission, droplet transmission, etc., which helps to assess the potential hazards of waste and the treatment measures that need to be taken; pathogen type refers to the type of pathogen that may be contained in the waste, such as bacteria, viruses, fungi, etc. Different types of pathogens may cause different health risks; the risk level of medical waste is classified according to its harmfulness to the environment and health, usually including high risk, medium risk and low risk. High-risk waste such as infectious waste needs to be strictly handled, while low-risk waste can be treated by conventional methods.

[0025] Preferably, multi-source feature collection of target medical waste refers to obtaining different types of information related to the target waste from multiple different sources or channels according to predetermined feature indicators, such as collecting data such as the generation time, processing requirements, storage location, historical processing records, etc. of the target medical waste, integrating information from different sources, and comprehensively understanding the characteristics of the target waste. The obtained multi-source data is used as the target feature indicator information, including all collected feature data (such as source, transmission route, pathogen type and risk level, etc.) and related data, and then an initial point code label is generated based on these feature data. Specifically, the initial point code label is a unique identifier of the target waste (such as a QR code, barcode, etc.), containing the feature information of the target waste (i.e., the target feature indicator information), and detailed data related to the waste can be easily obtained by scanning these labels. Each medical waste can be identified by a unique point code label (such as a QR code, barcode, etc.) to avoid confusion, and the use of a scanning device can quickly obtain detailed data of the waste, such as source, pathogen type, risk level, etc., which is convenient for tracking, management and recording, thereby realizing efficient management of medical waste from generation to treatment, and ensuring the traceability and safety of the waste.

[0026] Furthermore, the specific configuration of the target characteristic indicator information collection module 20 also includes collecting the target image of the target medical waste; identifying the text on the target image to obtain target label text information; analyzing the target label text information to determine the target source of the target medical waste; and adding the target source to the target characteristic indicator information.

[0027] Preferably, the image information of the target medical waste is obtained by some image acquisition device (such as a smart camera, mobile device or scanner, etc.), and the target image, that is, the image of the waste itself or the image information of the label, barcode, etc. on the waste is obtained. Then, the text in the image is recognized by using text recognition technology (OCR, optical character recognition), and the text information is extracted and converted into a data format that can be stored and processed. It may be the content written on the label, packaging, bag, etc. attached to the waste, which may include the type of medical waste, the source of the waste (such as a specific department or department), the production date or processing requirements, and the specific identification of the waste, etc. The text information output is analyzed to determine the source of the target medical waste. For example, the waste may come from different departments such as infectious diseases, emergency departments, operating rooms, etc., the type of waste (such as infectious waste, sharps waste, pharmaceutical waste, etc.) and the person responsible for waste treatment. After identifying and analyzing the source of the target waste, it is added to the characteristic indicator information of the target medical waste. The target characteristic indicator information is a collection of all key characteristic data of the waste, such as transmission route, pathogen type, risk level, etc. The relevant information of the target source is stored together with other characteristics (such as risk level, transmission route, etc.) for accurate tracking and management in the waste management system.

[0028] The waste clustering result obtaining module 30 is used to perform similarity clustering of the target medical wastes based on the target characteristic index information as a constraint to obtain waste clustering results.

[0029] Preferably, cluster analysis is used to classify multiple medical wastes based on key characteristic information of medical waste (i.e., target characteristic index information) to identify similar types of wastes. In medical waste management, clustering algorithms (K-means clustering, hierarchical clustering, DBSCAN, etc.) are used to classify similar wastes (such as wastes with similar sources, similar transmission routes, and similar pathogen types) into the same category by analyzing characteristic indicators (such as source, risk level, etc.) of different wastes. Specifically, the characteristic information (such as source, transmission route, etc.) of the target medical waste is used to classify similar wastes into the same category. The input data is waste type (such as infection pathway, risk level, etc.), and the similarity between each waste is calculated. A distance metric (such as Euclidean distance, Manhattan distance, etc.) is usually used to measure the similarity of the wastes. Based on the similarity information, a clustering algorithm is used to group similar wastes. For example, wastes with the same risk level and infection pathway may be clustered together. Finally, the waste clustering result is obtained, that is, multiple groups, each group represents a type of waste with similar characteristics, which helps to distinguish different types of waste, optimize the treatment process, improve the efficiency of resource allocation, and ensure that resources and equipment are reasonably allocated.

[0030] Furthermore, the specific configuration of the waste clustering result obtaining module 30 also includes obtaining first characteristic indicator information of a first medical waste; when the target characteristic indicator information and the first characteristic indicator information meet a predetermined clustering constraint, forming a first waste clustering cluster based on the target medical waste and the first medical waste; and forming the waste clustering result according to the first waste clustering cluster.

[0031] Preferably, the first medical waste is a type of medical waste existing in the system, and its first characteristic indicator information is obtained from the existing waste data, that is, a detailed characteristic description of the waste, including the type, source, infectiousness, risk level, etc. of the waste, and the characteristic information of the target medical waste is compared with the characteristic information of the first medical waste to determine whether a certain predetermined clustering constraint is met, wherein the predetermined clustering constraint is a preset condition, including that certain characteristics must be consistent, such as the target waste and the first waste come from the same department, the target waste and the first waste have similar risk levels, etc., and then the target waste that meets the clustering constraint and the first waste are aggregated into a waste clustering cluster to represent a class of medical waste with similar characteristics, for example, all wastes from the same department, with similar transmission routes and the same pathogen type will be aggregated into the same clustering cluster, and finally the waste clustering results are integrated and output to represent the grouping of all wastes that meet the clustering conditions, each clustering cluster contains some wastes with similar characteristics, which can be subsequently processed, managed or tracked according to their similarity, which is helpful for accurate classification and effective management of medical waste, and ensures the safety and efficiency of the waste treatment process.

[0032] Furthermore, the specific configuration of the waste clustering result acquisition module 30 also includes: obtaining a predetermined weight distribution of the predetermined characteristic indicator; analyzing and calculating the target characteristic indicator information and the first characteristic indicator information in sequence in combination with the predetermined weight distribution, and obtaining the target characteristic coefficient of the target medical waste and the first characteristic coefficient of the first medical waste respectively; comparing the target characteristic coefficient with the first characteristic coefficient to obtain a first characteristic deviation rate; introducing a clustering support evaluation function to analyze the first characteristic deviation rate to obtain a first clustering support; judging whether the first clustering support meets a predetermined support threshold; if not, adjusting and intervening in the first waste clustering cluster.

[0033] The specific configuration of the waste clustering result obtaining module 30 further includes that the expression of the clustering support evaluation function is as follows: ; in, refers to the first characteristic coefficient The target characteristic coefficient The first cluster support, k refers to the magnitude of the cluster support evaluation function, and , is the clustering decay factor, and .

[0034] Preferably, based on weighted characteristic indicators, clustering optimization is performed by analyzing the similarity of wastes to ensure that the clustering results are more reasonable and accurate, and predetermined weight distribution of predetermined characteristic indicators is obtained, wherein each characteristic indicator of medical waste (such as source, transmission route, pathogen type, risk level, etc.) has a predetermined weight, which reflects the importance of each feature in the clustering process, and is usually set based on the management needs of waste or the importance of waste treatment. For example, the transmission route may account for 30%, the pathogen type accounts for 40%, the risk level accounts for 30%, etc., and then according to the set weights, the characteristic information of the target medical waste and the first medical waste is analyzed and calculated, that is, each pair of characteristics (such as the transmission route of the target waste and the transmission route of the first waste) is compared, and each feature is weighted according to the set weight to obtain the characteristic coefficients of the two wastes, the target characteristic coefficient represents the comprehensive coefficient of all the characteristics of the target waste after weighted calculation, and the first characteristic coefficient represents the comprehensive coefficient of all the characteristics of the first waste after weighted calculation.

[0035] Preferably, after calculating the target characteristic coefficient and the first characteristic coefficient, the two coefficients are compared, and the first characteristic deviation rate is calculated to reflect the difference in weighted characteristic index between the target waste and the first waste. If the deviation rate is high, it means that there is a large difference in characteristics between the two wastes. Then, the first characteristic deviation rate is analyzed according to the cluster support evaluation function. Specifically, the cluster support evaluation function is used to evaluate the credibility and stability of the clustering results, and the support of the clustering is quantified based on the deviation rate to ensure that the clustering results can reflect the real waste characteristics. If the first characteristic deviation rate is very high, the cluster support may be low, which means that the characteristic difference between the target waste and the first waste is large. It may not be suitable to be clustered together; determine whether the calculated support of the first cluster exceeds the preset support threshold. The support threshold is a threshold set according to application requirements to determine whether two wastes can be reasonably clustered in the same cluster. For example, the support threshold can be set to 80%, indicating that if the cluster support is lower than 80%, it is considered that the cluster is not stable enough and the two wastes cannot be merged. If the support of the first cluster fails to reach the preset threshold, the first waste cluster is adjusted, for example, the members of the cluster are adjusted, the target waste is tried to be compared with other wastes, or new features are added to optimize the clustering results until a satisfactory support threshold is reached.

[0036] The predetermined waste disposal decision determination module 40 is used to determine the predetermined waste disposal decision of the target medical waste according to the waste clustering result matching.

[0037] Preferably, after clustering analysis of medical waste, the treatment plan for waste in each cluster is matched and determined based on the clustering results. Specifically, each cluster is analyzed based on the clustering results of the waste, and a corresponding treatment plan is formulated for each cluster. In combination with the characteristics of the cluster (such as the type of waste, risk level, etc.), the waste in the cluster is associated with a predetermined treatment decision. The predetermined waste treatment decision is a preset treatment plan based on the characteristics of medical waste. It is a treatment method set in advance based on information such as the category, characteristics, and risk level of the waste. For example, for waste clusters with strong infectiousness and high risks, special treatment methods such as high-temperature disinfection and chemical disinfection may be selected. For low-risk waste clusters, conventional harmless treatment (such as incineration or landfill) may be required. For certain medical waste (such as certain packaging materials), recycling and reuse treatment methods may be adopted. Through data-driven decision-making, the accuracy, efficiency and safety of waste management can be greatly improved.

[0038] The target processing image acquisition module 50 is used to call an intelligent image collector to dynamically monitor the treatment process of the target medical waste based on the predetermined waste treatment decision to obtain a target processing image.

[0039] Preferably, image acquisition technology (such as cameras, intelligent image sensors, etc.) is used to monitor and record the status of the target medical waste during the treatment process in real time, and generate corresponding image data, namely, the target processing image, to ensure that the waste treatment process complies with the predetermined treatment decision and can promptly detect possible problems or abnormalities. Specifically, the medical waste treatment process is efficiently monitored using an intelligent image collector, that is, the status of the target medical waste during the treatment process is tracked and monitored in real time to ensure that the treatment process is carried out in accordance with the predetermined waste treatment decision. Images or videos are obtained during the waste treatment process as target processing images. The target processing images may include images of the waste at various stages of the treatment process, whether the operator follows the correct steps, etc., thereby effectively improving the transparency, safety and efficiency of waste management.

[0040] The target person feature information matching module 60 is used to extract the target face slice in the target processed image based on face recognition technology, and match the target person feature information of the target person corresponding to the target face slice in the person database.

[0041] Preferably, in the process of medical waste treatment, face recognition technology is used to identify and track people in the processing image to obtain detailed information about the person, that is, based on the face recognition algorithm, the face part in the image is located and extracted by image processing technology to generate a facial feature area (i.e., target face slice). The target face slice refers to the facial area related to the operator extracted from the target image. The slice refers to the area in the image used to represent the face (i.e., facial feature area). Specifically, face recognition technology identifies and locates the face in the image by analyzing and processing image data, such as using a convolutional neural network (CNN) to extract facial feature points (such as eyes, nose, mouth, etc.) from the image and compare them, and finally extract an accurate face area; then the target person feature information of the target person corresponding to the target face slice is matched in the person database, wherein ,The personnel database refers to a database that stores the information of all personnel involved in the medical waste treatment, usually including each operator’s personal identity, job position, work record, processing capability, and other ,characteristic information related thereto. Specifically, after extracting the target face slice, the target face slice extracted from the processed image is ,compared with the facial features in the personnel database using a face recognition algorithm, such as calculating the distance and angle between the facial feature points to ,judge the accuracy of the match. When the matching facial features are found, they are ,associated to the corresponding operator, and the detailed information of the operator (such as name, position, work record, etc.) are ,extracted as the target personnel characteristic information, that is, the operator’s related personal information and ,identity characteristic data, which can improve the transparency of the medical waste treatment process, ensure ,the compliance of the operation process, reduce human errors and inappropriate ,behavior, and improve the safety of waste treatment.

[0042] Furthermore, the specific configuration of the target person feature information matching module 60 also includes: the target processing image includes multiple images of multiple processing stages, wherein the multiple processing stages include at least a packaging stage, a transportation stage, and a storage stage; extracting a first image from the multiple images, wherein the first image corresponds to a first processing stage; obtaining a first face slice in the first image using the principle of semantic recognition; matching the first person feature information of the first person corresponding to the first face slice in the person database; forming the target person based on the first person, and forming the target person feature information based on the first person feature information.

[0043] Preferably, waste treatment involves multiple stages, each stage generates a set of corresponding images, which together constitute the target treatment image, so the target treatment image includes multiple images of multiple treatment stages, and each stage may include different operations and management activities. The multiple treatment stages include at least a packaging stage, a transportation stage and a storage stage. Specifically, the packaging stage refers to the preliminary classification, packaging and labeling of waste in a medical institution or a waste generation point, and image acquisition is performed to record the waste packaging situation and whether the label is clear; the transportation stage refers to the transportation of waste from the generation point to the transfer station or treatment station, and image acquisition is performed to monitor the loading of the transport vehicle and the status of the waste during transportation; the storage stage refers to the temporary storage of waste in a specific area before final treatment, and image acquisition is performed to monitor the storage location, storage time, environmental conditions (such as temperature, humidity), etc.

[0044] Preferably, in the process of medical waste treatment, image recognition technology is used to extract and analyze personnel information from multiple images to ensure that the identity of the operator is associated with the treatment process and generate a tracking record. Specifically, one of the multiple images is arbitrarily selected as the first image, which records the first stage of waste treatment, and then the first face slice in the first image is obtained by using the semantic recognition principle, wherein semantic recognition refers to analyzing semantic information in an image through machine learning and computer vision technology, such as identifying objects, people, scenes, etc. in the image, and here refers to identifying the face area in the image, that is, extracting the area showing the operator's face (i.e., the face slice) from the first image, such as using a face detection algorithm (such as Haar cascade, deep learning, etc.) to locate and extract the face area in the image; then using face recognition technology , compare the first face slice extracted from the first image with all facial features in the personnel database, find the operator matching the face slice by calculating the similarity of the facial features, return the detailed information of the operator (such as name, position, etc.) as the first personnel feature information, finally form a target person according to the first person and form the target person feature information according to the first person feature information, that is, construct a complete identity information of the target person according to the matched first person feature information. For example, in addition to facial features, the target person feature information also includes other relevant data of the person (such as processing task records, operation qualifications, identity authentication, etc.), which is used to form a complete profile of the person in the waste treatment process, which helps to ensure that the identities of all personnel in the waste treatment process are confirmed, thereby improving the safety, transparency and compliance of waste management.

[0045] The target point code label acquisition module 70 is used to add the target person feature information to the initial point code label to obtain a target point code label, wherein the target point code label is used to track and manage the target medical waste.

[0046] Preferably, during the medical waste treatment process, the target person's characteristic information is added to the initial point code label, that is, the operator's identity information is associated with the tracking label of the waste itself (such as a QR code, barcode, etc.). Specifically, on the basis of the original initial point code label, the target person's characteristic information is added to form an updated target point code label, which can simultaneously record the basic information of the waste and the information of the specific operators involved in its treatment, providing a more comprehensive tracking tool that can not only track the source and treatment process of the waste, but also track the detailed identity of the operators involved in each link. The target point code label is used for the whole process management of the waste, and the current status of the waste, historical treatment records and operator information can be obtained in real time. Once any problem or accident occurs (such as improper treatment, leakage, dislocation, etc.), the specific waste and its treatment personnel can be quickly located, and the responsibilities of each operator can be clarified to ensure that there is a responsible person in each link, thereby improving operational transparency and enhancing the safety and compliance of the treatment process.

[0047] In the above, refer to Figure 1 The present invention describes in detail a medical waste tracking and management system combined with face recognition according to an embodiment of the present invention. Figure 2 A medical waste tracking and management method combined with face recognition according to an embodiment of the present invention is described.

[0048] Medical waste tracking and management methods combined with face recognition, such as Figure 2 As shown, the method includes: obtaining target medical waste, wherein the target medical waste refers to any waste that meets the predetermined waste constraint; calling the predetermined characteristic index to perform multi-source feature collection on the target medical waste to obtain the target characteristic indicator information, and generating the initial point code label of the target characteristic indicator information; performing similarity clustering of the target medical waste with the target characteristic indicator information as a constraint to obtain the waste clustering result; determining the predetermined waste treatment decision of the target medical waste according to the matching of the waste clustering result; calling the intelligent image collector to dynamically monitor the treatment process of the target medical waste based on the predetermined waste treatment decision to obtain the target treatment image; extracting the target face slice in the target treatment image based on the face recognition technology, and matching the target personnel feature information of the target personnel corresponding to the target face slice in the personnel database; adding the target personnel feature information to the initial point code label to obtain the target point code label, wherein the target point code label is used to track and manage the target medical waste.

[0049] Although the present application makes various references to certain modules in the system according to the embodiments of the present application, any number of different modules may be used and run on the user terminal and / or server, and the various units and modules included are only divided according to functional logic, but are not limited to the above division, as long as the corresponding functions can be achieved; in addition, the specific names of the functional units are only for the convenience of distinguishing each other, and are not used to limit the scope of protection of the present invention.

[0050] The above specific implementations do not constitute a limitation on the protection scope of this application. It should be understood by those skilled in the art that various modifications, combinations and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions and improvements made within the spirit and principles of this application should be included in the protection scope of this application.

Claims

1. The medical waste tracking and management system combined with face recognition is characterized by: The system comprises: A target medical waste acquisition module, used to acquire target medical waste, wherein the target medical waste refers to any waste that meets the predetermined waste constraints; A target characteristic indicator information collection module is used to retrieve predetermined characteristic indicators to perform multi-source characteristic collection on the target medical waste to obtain target characteristic indicator information, and generate an initial point code label for the target characteristic indicator information; A waste clustering result obtaining module, used to perform similarity clustering of the target medical wastes with the target characteristic index information as a constraint to obtain a waste clustering result; A predetermined waste disposal decision determination module, used to determine the predetermined waste disposal decision of the target medical waste according to the waste clustering result matching; A target processing image acquisition module is used to call an intelligent image collector to dynamically monitor the treatment process of the target medical waste based on the predetermined waste treatment decision to obtain a target processing image; A target person feature information matching module is used to extract a target face slice in the target processed image based on face recognition technology, and match the target person feature information of the target person corresponding to the target face slice in a person database; The target point code label acquisition module is used to add the target person feature information to the initial point code label to obtain the target point code label, wherein the target point code label is used to track and manage the target medical waste.

2. According to claim 1, the medical waste tracking and management system combined with face recognition is characterized in that: The predetermined characteristic indicators include at least source, route of infection, pathogen type and risk level.

3. According to claim 2, the medical waste tracking and management system combined with face recognition is characterized in that: The target characteristic indicator information collection module performs the following steps: Acquiring a target image of the target medical waste; Recognize the text on the target image to obtain target label text information; Analyzing the target label text information to determine the target source of the target medical waste; The target source is added to the target feature indicator information.

4. The medical waste tracking and management system combined with face recognition according to claim 1 is characterized in that: The waste clustering result obtaining module performs the following steps: Obtaining first characteristic indicator information of first medical waste; When the target characteristic index information and the first characteristic index information meet the predetermined clustering constraint, forming a first waste cluster based on the target medical waste and the first medical waste; The waste clustering result is formed according to the first waste clustering clusters.

5. The medical waste tracking and management system combined with face recognition according to claim 4 is characterized in that: The waste clustering result obtaining module further includes the following steps: Obtaining a predetermined weight distribution of the predetermined characteristic indicators; In combination with the predetermined weight distribution, the target characteristic index information and the first characteristic index information are analyzed and calculated in sequence to obtain a target characteristic coefficient of the target medical waste and a first characteristic coefficient of the first medical waste respectively; Comparing the target characteristic coefficient with the first characteristic coefficient to obtain a first characteristic deviation rate; Introducing a cluster support evaluation function to analyze the first feature deviation rate to obtain a first cluster support; Determining whether the support of the first clustering reaches a predetermined support threshold; If not satisfied, the first waste cluster is adjusted and intervened.

6. The medical waste tracking and management system combined with face recognition according to claim 5 is characterized in that: The expression of the cluster support evaluation function is as follows: ； in, refers to the first characteristic coefficient The target characteristic coefficient The first cluster support, k refers to the magnitude of the cluster support evaluation function, and , is the clustering decay factor, and .

7. The medical waste tracking and management system combined with face recognition according to claim 1 is characterized in that: The target person feature information matching module performs the following steps: Extracting a first image from the plurality of images, the first image corresponding to a first processing stage; Acquire a first face slice in the first image by using a semantic recognition principle; Matching first person feature information of a first person corresponding to the first face slice in the person database; The target personnel are formed based on the first personnel, and the target personnel characteristic information is formed based on the first personnel characteristic information.

8. A medical waste tracking and management method combined with face recognition, the method being applied to the medical waste tracking and management system combined with face recognition according to any one of claims 1 to 7, the method comprising: Obtaining target medical waste, where the target medical waste refers to any waste that meets a predetermined waste constraint; Retrieving predetermined characteristic indicators to collect multi-source characteristics of the target medical waste to obtain target characteristic indicator information, and generating an initial point code label for the target characteristic indicator information; Perform similarity clustering of the target medical waste using the target characteristic index information as a constraint to obtain a waste clustering result; Determining a predetermined waste disposal decision for the target medical waste according to the waste clustering result matching; Calling an intelligent image collector to dynamically monitor the treatment process of the target medical waste based on the predetermined waste treatment decision to obtain a target treatment image; Extracting a target face slice in the target processed image based on face recognition technology, and matching target person feature information of a target person corresponding to the target face slice in a person database; The target person characteristic information is added to the initial point code label to obtain a target point code label, wherein the target point code label is used to track and manage the target medical waste.

Images