A medical waste supervision node hidden danger intelligent early warning method and system

Abstract

This invention discloses an intelligent early warning method and system for potential hazards at key nodes in the medical waste supervision process. Based on a fuzzy inference rule base, the invention utilizes the Mamdani inference algorithm to perform fuzzy inference on the fuzzy sets of information risk and physical risk for each node, generating a comprehensive risk fuzzy set for each node. The centroid method is then used to defuzzify the comprehensive risk fuzzy set, yielding a comprehensive risk value for each node. Based on the comprehensive risk value, early warning levels are assigned, and finally, the early warning results and risk sources for each node in the medical waste supervision process are output. This invention comprehensively considers risks at both the physical and information layers, improving the accuracy of early warnings for nodes in the medical waste supervision process.

Description

Technical Field

[0001] This invention relates to the field of medical waste management technology, specifically to an intelligent early warning method and system for potential hazards at medical waste monitoring nodes. Background Technology

[0002] Medical waste refers to waste generated during medical, preventive, healthcare, and related activities that possess direct or indirect infectiousness, toxicity, or other hazards. This type of waste is highly hazardous, and risks such as loss, leakage, or illegal dumping may occur at every stage from generation, sorting, collection, temporary storage, and transportation to final disposal. The comprehensive supervision of medical waste throughout its entire lifecycle is directly related to public health safety and environmental safety.

[0003] The medical waste supervision process involves several key nodes, each of which may face dual risks: informational and physical risks. Informational risks mainly manifest as errors in the synchronization of electronic manifests, delays in information transmission, and incomplete record keeping, directly affecting the traceability and accuracy of supervision. Physical risks mainly manifest as medical waste leakage, container damage, illegal transportation, substandard storage environments, and non-compliant disposal processes, which can easily lead to serious safety incidents such as environmental pollution and cross-infection.

[0004] Existing medical waste regulatory risk warning technologies mostly rely on single risk indicators for assessment at the physical level, resulting in low warning accuracy, delayed response, and unclear risk source identification. Simple threshold judgments can only provide warnings for a single risk indicator and cannot comprehensively consider the coupling effect of information risk and physical risk. This easily leads to false or missed warnings, and it cannot clearly identify the specific risk source behind the warning, making it difficult for regulatory personnel to carry out targeted rectification and failing to meet the needs of refined and intelligent medical waste supervision. Summary of the Invention

[0005] The purpose of this invention is to provide an intelligent early warning method and system for potential risks at medical waste supervision nodes, which considers both information risks and physical risks in the medical waste supervision process to provide risk warnings for medical waste supervision nodes.

[0006] To solve the above-mentioned technical problems, the present invention provides the following technical solution: A method for intelligent early warning of potential hazards at medical waste monitoring nodes, comprising the following steps: Data from each medical waste management process node is collected in the physical and information domains to obtain the first physical data and second information data for each node. The medical waste management process nodes include generation nodes, collection nodes, temporary storage nodes, transfer nodes, and disposal nodes. Calculate the risk confidence levels of the first physical data and the second information data respectively; Based on the risk confidence of the physical and information data of each medical waste supervision node, the membership degree is calculated and fuzzified to generate the physical risk fuzzy set and information risk fuzzy set of the corresponding node. Construct a fuzzy reasoning rule base; Based on the fuzzy reasoning rule base, the Mamdani reasoning algorithm is used to perform reasoning on the information risk fuzzy set and physical risk fuzzy set of each node, and output the node comprehensive risk fuzzy set; The fuzzy set of the node's comprehensive risk is defuzzified using the centroid method to obtain the node's comprehensive risk value. Based on the node's comprehensive risk value, the corresponding early warning level is divided, and the early warning and risk source of the node in the medical waste supervision process are output.

[0007] This invention performs source tracing analysis of risks based on the contribution components of the output fuzzy set generated during the Mamdani inference process. The product of the activation intensity and the area of ​​the corresponding fuzzy conclusion subset is the contribution component of the output fuzzy set corresponding to a single rule. The total contribution of rules triggered by medium or high risks of information or physical risks to the final fuzzy output is calculated. The main source of risk is determined by the contribution degree.

[0008] Meanwhile, if the overall risk value of the node is less than the first-level safety threshold, the medical waste supervision process node is considered to be operating safely; if the overall risk value of the node is greater than or equal to the first-level safety threshold and less than the second-level safety threshold, the medical waste supervision process node is considered to be under first-level warning; if the overall risk value of the node is greater than or equal to the second-level safety threshold, the medical waste supervision process node is considered to be under second-level warning.

[0009] Based on the warning level, the corresponding audible and visual alarms, SMS messages, or telephone warnings will be triggered immediately and simultaneously pushed to the supervisory person in charge and emergency response personnel; warning records will also be generated and regularly summarized and pushed to the regulatory platform to remind rectification.

[0010] According to the above technical solution, the first physical data of the medical waste supervision process node includes medical waste weight, RFID tag identification status, temporary storage room temperature and humidity, handover verification status, packaging integrity, medical waste identification result, and leakage sensor data. The second information data of the medical waste supervision process nodes includes electronic manifest synchronization status, data upload delay, interface packet loss rate, platform response time, operation log, access permission records, and process node dwell time.

[0011] Due to functional differences at each node of the medical waste supervision process, the first physical data collected at each node may not be entirely consistent with the second information. Therefore, the first physical data for each medical waste supervision process node includes one or more of the following: medical waste weight, RFID tag identification status, temporary storage room temperature and humidity, handover verification status, packaging integrity, medical waste identification result, and leakage sensor data. Similarly, the second information data for each medical waste supervision process node includes one or more of the following: electronic manifest synchronization status, data upload delay, interface packet loss rate, platform response time, operation log, access permission records, and process node dwell time.

[0012] According to the above technical solution, the first physical data and the second information data are divided into binary data and non-binary data; the binary data includes RFID tag identification status, handover verification status, packaging integrity, medical waste identification results, operation logs and access records; The non-binary data includes the weight of medical waste, leakage sensor data, temperature and humidity of the temporary storage room, data upload delay, interface packet loss rate, platform response time, and the dwell time of the process nodes.

[0013] Because the first physical data and the second information data contain binary data, membership degrees cannot be directly calculated. Therefore, it is necessary to calculate the risk confidence level first and then calculate the membership degree. One purpose is to unify the dimensions, and another is to bind the credibility of risk with the degree of fuzzy membership, thereby solving the deficiency of traditional fuzzy assessment that only considers degree and not credibility. This effectively improves the scientificity, robustness, and accuracy of risk assessment, reduces subjective bias and data noise interference, and makes risk classification and decision-making more reasonable, interpretable, and adaptable to complex and uncertain scenarios.

[0014] According to the above technical solution, the steps for calculating the risk confidence of non-binary data include: Obtain historical normal data for non-binary data; calculate the median Q2, Q1, Q3, and interquartile range. ; The interquartile range of the non-binary data is calculated based on historical normal data to determine the corresponding normal value range. ; Anomaly scores are obtained based on the deviation of the non-binary data from the median and the ratio of the interquartile range. ; The outlier scores are normalized to the [0,1] interval and then used as the risk confidence level of the corresponding non-binary data.

[0015] Risk confidence level is used to characterize the current degree of abnormal risk of the indicator. The closer the risk confidence level is to 1, the higher the degree of risk.

[0016] According to the above technical solution, if the binary data judgment result is an abnormal state, then the risk confidence level of the binary data is 1; otherwise, the risk confidence level is 0.

[0017] According to the above technical solution, the risk confidence of the first physical data and the second information data is converted into membership degree using a triangular membership function or a trapezoidal membership function.

[0018] For the binary anomaly confidence of RFID tag identification status, handover verification status, packaging integrity, medical waste identification results, and leakage sensor data, the membership degree can also be transformed using triangular membership function or trapezoidal membership function; where the anomaly confidence is 0, it is transformed into anomaly membership degree 0, and the anomaly confidence is 1, it is transformed into anomaly membership degree 1, thus obtaining anomaly membership degree normalized to the [0,1] interval.

[0019] The risk confidence levels of the first physical data and the second information data are fuzzyened into low risk, medium-high risk, and high risk.

[0020] This process involves establishing a fuzzy inference rule base. The Mamdani inference algorithm is used to match rules in the fuzzy rule base, calculating the activation strength of each activation rule, and obtaining a subset of fuzzy conclusions for each rule through truncation operations. The total output fuzzy set is then obtained through aggregation by taking the largest value. Finally, the centroid method is used for defuzzification to obtain the comprehensive risk value of the node. This process realizes the transformation from fuzzy risk information to accurate assessment results, providing reliable methodological support for the quantitative assessment of node risk.

[0021] According to the above technical solution, the comprehensive risk values ​​of each node in the medical waste supervision process are weighted and integrated to obtain the comprehensive risk value of the entire medical waste supervision process. Based on the comprehensive risk value of the entire medical waste supervision process, corresponding early warning levels are divided, and a comprehensive risk assessment of the medical waste supervision process is carried out.

[0022] This includes a technical solution, a medical waste supervision node hidden danger intelligent early warning system, comprising: The data acquisition module is used to collect data from each medical waste supervision process node in the physical domain and information domain, and obtain the first physical data and second information data of the corresponding node respectively; the medical waste supervision process nodes include generation node, collection node, temporary storage node, transfer node, and disposal node. A confidence level determination module is used to calculate the risk confidence levels of the first physical data and the second information data, respectively. The membership determination module calculates and fuzzifies the membership degree based on the risk confidence of the physical and information data of each medical waste supervision node, generating the physical risk fuzzy set and information risk fuzzy set of the corresponding node. The rule base building module is used to construct the fuzzy inference rule base; The fuzzy reasoning module, based on the fuzzy reasoning rule base, uses the Mamdani reasoning algorithm to perform reasoning on the information risk fuzzy set and physical risk fuzzy set of each node, and outputs the node comprehensive risk fuzzy set; The early warning module uses the centroid method to defuzzify the fuzzy set of the node's comprehensive risk to obtain the node's comprehensive risk value. Based on the node's comprehensive risk value, it classifies the corresponding early warning level and outputs the early warning and risk source of the node in the medical waste supervision process.

[0023] The invention includes a technical solution that provides an electronic device comprising: one or more processors; and a storage device for storing one or more programs, wherein when the one or more programs are executed by the one or more processors, the one or more processors execute a method for intelligent early warning of potential hazards at medical waste monitoring nodes as described in any of the above technical solutions.

[0024] One technical solution includes a storage medium storing at least one instruction, which is loaded and executed by a processor to implement a method for intelligent early warning of potential hazards at medical waste monitoring nodes as described in any of the above technical solutions.

[0025] Compared with existing technologies, the beneficial effects achieved by this invention are as follows: Based on a fuzzy inference rule base, this invention utilizes the Mamdani inference algorithm to perform fuzzy inference on the information risk fuzzy set and physical risk fuzzy set of each node, generating a node comprehensive risk fuzzy set; the centroid method is used to perform defuzzification processing on the comprehensive risk fuzzy set to obtain the node comprehensive risk value; based on the comprehensive risk value, early warning levels are divided, and finally, the early warning results and risk sources of each node in the medical waste supervision process are output. This invention covers all regulatory nodes of medical waste generation, temporary storage, transfer, and disposal, realizing real-time collection and fusion of multi-source data, breaking down data silos, and comprehensively considering risks at both the physical and information levels, making the early warning of the medical waste supervision process more comprehensive and accurate, enabling timely preventive measures. Furthermore, the identification of risk sources can better optimize the medical waste supervision process. Attached Figure Description

[0026] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used in conjunction with embodiments of the invention to explain the invention and do not constitute a limitation thereof. In the drawings: Figure 1 This is a flowchart illustrating the steps of an intelligent early warning method for potential hazards at medical waste monitoring nodes according to the present invention. Figure 2 Flowchart for calculating risk confidence of non-binary data. Detailed Implementation

[0027] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0028] Based on the actual process of medical waste circulation in hospitals, this invention provides an intelligent early warning method for potential hazards at key medical waste supervision nodes, enabling early warning at the generation, collection, temporary storage, transfer, and disposal nodes of the medical waste supervision process. Specific steps include (…). Figure 1 )include: S1. Collect data from each node in the medical waste management process in the physical and information domains, obtaining the first physical data and second information data for each node, specifically: The first physical data collected by the generating node includes the medical waste identification results; the second information data collected by the generating node includes data upload delay, interface packet loss rate, platform response time, operation log, permission access record, and process node dwell time.

[0029] The first physical data collected by the collection nodes includes the weight of medical waste, packaging integrity, handover verification status, leakage sensor data, and RFID tag identification status; the second information data collected by the collection nodes includes the synchronization status of electronic manifests, data upload delay, interface packet loss rate, platform response time, operation logs, access permission records, and process node dwell time.

[0030] The first physical data collected by the temporary storage node includes the temperature and humidity of the temporary storage room, leakage sensor data, and packaging integrity; the second information data collected by the temporary storage node includes the synchronization status of electronic manifests, data upload delay, interface packet loss rate, platform response time, operation logs, access permission records, and process node dwell time.

[0031] The first physical data collected by the transfer node includes vehicle location data, leakage sensor data, packaging integrity, and handover verification status; the second information data collected by the transfer node includes electronic manifest synchronization status, data upload delay, interface packet loss rate, platform response time, operation logs, access permission records, and process node dwell time.

[0032] The first physical data collected at the disposal node includes the weight of medical waste, RFID tag identification status, packaging integrity, handover verification status, leakage sensor data, medical waste identification results, and temperature and humidity of the temporary storage room. The second information data collected at the disposal node includes the synchronization status of electronic manifests, data upload latency, interface packet loss rate, platform response time, operation logs, access permission records, and process node dwell time.

[0033] The primary physical data and secondary information data for each medical waste management process node are preprocessed, such as removing abnormal data and supplementing missing data. Furthermore, the primary physical data and secondary information data for each node are time-aligned.

[0034] S2. Calculate the risk confidence levels for the first physical data and the second information data respectively. Since some data in both the first physical data and the second information data are binary, different methods are used to calculate the corresponding risk confidence levels based on the data type.

[0035] RFID tag identification status, handover verification status, packaging integrity, medical waste identification results, operation logs, and access records can all be considered as binary data. When the RFID tag identification status is "identification passed," the corresponding risk confidence level is 0; if the RFID tag identification status is "identification failed," the corresponding risk confidence level is 1. When the handover verification status is "verification passed," the corresponding risk confidence level is 0; when the handover verification status is "verification failed," the corresponding risk confidence level is 1. If the packaging is intact, the corresponding risk confidence level is 0; if the packaging is incomplete, the corresponding risk confidence level is 1. If the medical waste identification result is "passed," the corresponding risk confidence level is 0; if the medical waste is identified incorrectly, the corresponding risk confidence level is 1. If access is normal, the corresponding risk confidence level is 0; if access is abnormal, the corresponding risk confidence level is 1. If the operation log is normal, the corresponding risk confidence level is 0; if the operation log is abnormal, the corresponding risk confidence level is 1.

[0036] For non-binary data such as medical waste weight, leakage sensor data, temporary storage room temperature and humidity, data upload latency, interface packet loss rate, platform response time, and process node dwell time, the interquartile range is used to determine the risk confidence level. Figure 2 Specifically: First, the corresponding interquartile range is calculated based on the historical normal data of non-binary data to determine the corresponding normal value range; then, based on the deviation of the current non-binary data from the median and the ratio of the interquartile range, the anomaly score is obtained, and after normalization to the [0,1] interval, it is used as the risk confidence of the corresponding non-binary data.

[0037] For example, regarding the weight of medical waste, 15 historical normal data points were obtained: 2.1, 2.3, 2.5, 2.7, 2.8, 3.0, 3.1, 3.3, 3.4, 3.6, 3.7, 3.9, 4.1, 4.3, 4.5; the current weight of medical waste is 7.0. After sorting, Q1 takes the value 2.7, the median Q2 takes the value 3.3, and Q3 takes the value 3.9; then... The normal range for the weight of medical waste is: ; That is, the abnormal score of medical waste is Abnormal scores for medical waste are normalized to the [0,1] interval. Specifically, a maximum score limit of 5 can be set (scores exceeding this limit are also counted as 5). The risk confidence level for medical waste is then calculated as follows: .

[0038] S3. Calculate the membership degree of the first physical data and second information data for each medical waste supervision process node using either a triangular membership function or a trapezoidal membership function, respectively, based on their risk confidence levels. Simultaneously, fuzzify the risk confidence levels of the first physical data and second information data into low risk, medium-high risk, and high risk categories.

[0039] Among them, the binary anomaly confidence of RFID tag identification status, handover verification status, packaging integrity, medical waste identification results, and leakage sensor data is converted into anomaly membership degree 0 when the anomaly confidence degree is 0, and into anomaly membership degree 1 when the anomaly confidence degree is 1, so as to obtain anomaly membership degree normalized to the interval [0,1].

[0040] S4. Fuzzify the membership degrees of all first physical data corresponding to the generation, collection, temporary storage, transfer, and disposal nodes of the medical waste management process to obtain the physical risk fuzzy set of that node. Similarly, fuzzify the membership degrees of all second information data corresponding to the generation, collection, temporary storage, transfer, and disposal nodes of the medical waste management process to obtain the information risk fuzzy set of that node. For example, the medical waste identification result after fuzzification of the generation node is high risk.

[0041] S5. Construct a fuzzy reasoning rule base. For example, if the first physical data is low risk (L) and the second information data is low risk (L), then the warning level is normal; if the first physical data is low risk (L) and the second information data is medium risk (M), then the warning level is low risk; if the first physical data is low risk (L) and the second information data is high risk (H), then the warning level is medium-high risk; if the first physical data is medium risk (M) and the second information data is low risk (L), then the warning level is low risk; if the first physical data is medium risk (M .... If the data is of medium risk (M), the warning level is medium-high risk; if the first physical data is of medium risk (M) and the second information data is of high risk (H), the warning level is high risk; if the first physical data is of high risk (H) and the second information data is of low risk (L), the warning level is medium-high risk; if the first physical data is of high risk (H) and the second information data is of medium risk (M), the warning level is high risk; if the first physical data is of high risk (H) and the second information data is of high risk (H), the warning level is high risk.

[0042] S6. Based on the fuzzy inference rule base, use the Mamdani inference algorithm to perform inference on the information risk fuzzy set and physical risk fuzzy set of each node, and output the node comprehensive risk fuzzy set; use the centroid method to defuzzify the node comprehensive risk fuzzy set to obtain the node comprehensive risk value, divide the corresponding warning level according to the node comprehensive risk value, and output the warning and risk source of this medical waste supervision process node.

[0043] If the node comprehensive risk value is less than the first-level safety threshold of 0.3, then this medical waste supervision process node is operating safely; if the node comprehensive risk value is greater than or equal to the first-level safety threshold of 0.3 and less than the second-level safety threshold of 0.7, then this medical waste supervision process node is at the first-level warning; if the node comprehensive risk value is greater than or equal to the second-level safety threshold of 0.7, then this medical waste supervision process node is at the second-level warning. Immediately trigger corresponding acoustic and optical alarms, text messages or calls and other warnings according to the warning level, and push them to the supervision responsible person and emergency disposal personnel synchronously; and generate warning records, and regularly summarize and push them to the supervision platform to remind of rectification.

[0044] At the same time, it is possible to perform source tracing analysis of the hazard source based on the output fuzzy set contribution components generated during the Mamdani inference process. Rectify according to the hazard source.

[0045] S7. Perform weighted fusion on the node comprehensive risk values of each medical waste supervision process node to obtain the comprehensive risk value of the entire medical waste supervision process, divide the corresponding warning level according to the comprehensive risk value of the entire medical waste supervision process, and perform comprehensive risk assessment of the medical waste supervision process.

[0046] It should be noted that in this article, relational terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the term "comprising", "including" or any other variant thereof is intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements not only includes those elements, but also includes other elements not expressly listed, or also includes elements inherent to such process, method, article or device.

[0047] Finally, it should be noted that the above are only the preferred embodiments of the present invention and are not used to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, for those skilled in the art, they can still modify the technical solutions recorded in the foregoing embodiments, or perform equivalent replacements on some of the technical features. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present invention shall be included within the protection scope of the present invention.

Claims

1. A method for intelligent early warning of potential hazards at medical waste monitoring nodes, characterized in that, The steps include: Data from each medical waste management process node is collected in the physical and information domains to obtain the first physical data and second information data for the corresponding node; the medical waste management process nodes include generation nodes, collection nodes, temporary storage nodes, transfer nodes, and disposal nodes; Calculate the risk confidence levels of the first physical data and the second information data respectively; Based on the risk confidence of the physical and information data of each medical waste supervision node, the membership degree is calculated and fuzzified to generate the physical risk fuzzy set and information risk fuzzy set of the corresponding node. Construct a fuzzy reasoning rule base; Based on the fuzzy reasoning rule base, the Mamdani reasoning algorithm is used to perform reasoning on the information risk fuzzy set and physical risk fuzzy set of each node, and output the node comprehensive risk fuzzy set; The fuzzy set of the node's comprehensive risk is defuzzified using the centroid method to obtain the node's comprehensive risk value. Based on the node's comprehensive risk value, the corresponding early warning level is divided, and the early warning and risk source of the node in the medical waste supervision process are output.

2. The intelligent early warning method for potential hazards at medical waste monitoring nodes according to claim 1, characterized in that, The first physical data of the medical waste supervision process nodes includes medical waste weight, RFID tag identification status, temporary storage room temperature and humidity, handover verification status, packaging integrity, medical waste identification results, and leakage sensor data. The second information data of the medical waste supervision process nodes includes electronic manifest synchronization status, data upload delay, interface packet loss rate, platform response time, operation log, access permission records, and process node dwell time.

3. The intelligent early warning method for potential hazards at medical waste monitoring nodes according to claim 2, characterized in that, The first physical data and the second information data are divided into binary data and non-binary data; the binary data includes RFID tag identification status, handover verification status, packaging integrity, medical waste identification results, operation logs, and the access permission records. The non-binary data includes the weight of medical waste, leakage sensor data, temperature and humidity of the temporary storage room, data upload delay, interface packet loss rate, platform response time, and the dwell time of the process nodes.

4. The intelligent early warning method for potential hazards at medical waste monitoring nodes according to claim 3, characterized in that, The steps for calculating the risk confidence of non-binary data include: Retrieve historical normal data for non-binary data; Calculate the interquartile range of the non-binary data based on historical normal data to determine the corresponding normal value range; Anomaly scores are obtained based on the degree of deviation of the non-binary data from the median and the interquartile range ratio. The outlier scores are normalized to the [0,1] interval and then used as the risk confidence level of the corresponding non-binary data.

5. The intelligent early warning method for potential hazards at medical waste monitoring nodes according to claim 3, characterized in that, If the binary data indicates an abnormal state, the risk confidence level of the binary data is 1; otherwise, the risk confidence level is 0.

6. The intelligent early warning method for potential hazards at medical waste monitoring nodes according to claim 1, characterized in that, The risk confidence of the first physical data and the second information data is converted into membership degree using a triangular membership function or a trapezoidal membership function.

7. The intelligent early warning method for potential hazards at medical waste monitoring nodes according to claim 1, characterized in that, The comprehensive risk values ​​of each node in the medical waste supervision process are weighted and integrated to obtain the comprehensive risk value of the entire medical waste supervision process. Based on the comprehensive risk value of the entire medical waste supervision process, corresponding early warning levels are divided, and a comprehensive risk assessment of the medical waste supervision process is carried out.

8. A smart early warning system for potential hazards at medical waste monitoring nodes, characterized in that, include: The data acquisition module is used to collect data from each node of the medical waste supervision process in the physical and information domains, and obtain the first physical data and the second information data of the corresponding node respectively. The medical waste supervision process includes the following nodes: generation node, collection node, temporary storage node, transfer node, and disposal node. A confidence level determination module is used to calculate the risk confidence levels of the first physical data and the second information data, respectively. The membership determination module calculates and fuzzifies the membership degree based on the risk confidence of the physical and information data of each medical waste supervision node, generating the physical risk fuzzy set and information risk fuzzy set of the corresponding node. The rule base building module is used to construct the fuzzy inference rule base; The fuzzy reasoning module, based on the fuzzy reasoning rule base, uses the Mamdani reasoning algorithm to perform reasoning on the information risk fuzzy set and physical risk fuzzy set of each node, and outputs the node comprehensive risk fuzzy set. The early warning module uses the centroid method to defuzzify the fuzzy set of the node's comprehensive risk to obtain the node's comprehensive risk value. Based on the node's comprehensive risk value, it classifies the corresponding early warning level and outputs the early warning and risk source of the node in the medical waste supervision process.

9. An electronic device, characterized in that, include: One or more processors; A storage device for storing one or more programs, wherein when the one or more programs are executed by the one or more processors, the one or more processors cause the one or more processors to perform the method according to any one of claims 1-7.

10. A storage medium, characterized in that, The storage medium stores at least one instruction, which is loaded and executed by a processor to implement the method as described in any one of claims 1-7.

Images