Radioactive operation safety assessment and warning system for nuclear medicine medication nursing

Abstract

The invention provides a radioactive operation safety assessment and warning system for nuclear medicine medication nursing, and the system comprises a data obtaining and processing module which is used for deploying a space monitoring network and detecting radiation data; the real-time radiation monitoring module is used for collecting operation behavior data and dose data in real time and carrying out abnormal operation identification and risk scoring; the intelligent prediction and evaluation module is used for constructing a risk prediction model and carrying out risk prediction on the real-time data to obtain a predicted risk level; the operation safety evaluation module is used for automatically calculating the safety score of the operation and generating a safety evaluation report; and the diversified intelligent warning module is used for calculating a comprehensive score and carrying out graded sound-light alarm. Through efficient and real-time data monitoring and intelligent and accurate risk assessment and operation safety assessment, the risk of radioactive exposure in nuclear medicine medication nursing can be effectively reduced, the safety of medical staff and patients is guaranteed, and the scientificity and normalization of medical operation are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of nuclear medicine nursing, and in particular to a radioactive operation safety assessment and warning system for nuclear medicine medication nursing. Background Art

[0002] Radiological operation safety assessment and alerting involves systematically evaluating the operational risks of radioactive materials involved in nuclear medicine medication administration, and issuing timely safety alerts through technical means to prevent radioactive contamination, overexposure, and related accidents, thereby ensuring the safety of patients, medical staff, and the environment. Currently, safety assessment and alerting for radiological operations are mostly based on radiation monitoring equipment, operational procedures, and risk assessment models. Existing technologies are fragmented and limited in scope, and suffer from the following issues: 1) Limited coverage: They often focus solely on dose measurement or environmental radiation, lacking comprehensive operational safety assessments. 2) Low levels of intelligence and automation: They lack intelligent risk prediction and analysis that integrates artificial intelligence and big data. 3) Single alert methods: They are mostly audio or light alarms with fixed thresholds, lacking flexibility and customization. 4) Manual operation: Operational procedures and safety checks are often manually executed, which can easily lead to human oversight. 5) Inadequate data integration and tracking: Dispersed operational data makes it difficult to maintain a complete operational and safety log, hindering subsequent assessment and verification. Summary of the Invention

[0003] In order to overcome the shortcomings of the existing technology, the purpose of the present invention is to provide a radioactive operation safety assessment and warning system for nuclear medicine medication nursing. Through efficient and real-time data monitoring and intelligent and accurate risk assessment and operation safety assessment, it can effectively reduce the risk of radioactive exposure in nuclear medicine medication nursing, ensure the safety of medical staff and patients, and improve the scientificity and standardization of medical operations.

[0004] To achieve the above objectives, the present invention provides the following solution: a radioactive operation safety assessment and warning system for nuclear medicine medication nursing, comprising:

[0005] The data acquisition and processing module is used to deploy a space monitoring network in the nuclear medicine medication operation area, use the space monitoring network to monitor the changes of radioactive materials in real time, obtain radiation data, and use a distributed storage network to store and manage the radiation data;

[0006] The real-time radiation monitoring module is used to collect the operation behavior data of medical staff in real time using smart wearable devices, and deploy RFID readers and touch screens on the operating table to read the dose data in real time. Based on the operation behavior data and the dose data, anomalies are identified to obtain abnormal operation behaviors. Based on the operation behavior data and the radiation data, risk scoring is performed to obtain high-risk operation points and complete risk assessment;

[0007] An intelligent prediction and assessment module is used to build a risk prediction model, use the risk prediction model to predict the risk of real-time data, and obtain the predicted risk level;

[0008] An operation safety assessment module is used to build a safety standard rule base, automatically calculate the safety score of the operation based on the safety standard rule base and use a preset safety assessment engine to generate a safety assessment report;

[0009] A diversified intelligent warning module is used to integrate the safety score, the risk score and the predicted risk level into a comprehensive score, and then perform graded sound and light alarms based on the comprehensive score;

[0010] Among them, the data acquisition and processing module, the real-time radiation monitoring module, the intelligent prediction and evaluation module, the operation safety evaluation module, and the diversified intelligent warning module are interconnected.

[0011] Optionally, the data acquisition and processing module includes:

[0012] Radiation sensor units are used to deploy semiconductor radiation sensors within the nuclear medicine administration area to detect changes in radioactive substances, and to design an automatic sensor calibration mechanism within the semiconductor radiation sensors to complete the deployment of the operation space monitoring network;

[0013] A data initial processing unit is used to transmit the monitoring data of the semiconductor radiation sensor to the edge computing unit in real time, perform preliminary processing and data integration of the monitoring data to obtain a real-time monitoring data set, and then transmit the real-time monitoring data set to the cloud central control system for processing;

[0014] A data cleaning unit is used to perform data cleaning, missing value filling, outlier extraction and time series synchronization operations on the real-time monitoring data set based on the cloud central control system, and then combine the operation behavior data and environmental sensor data to cross-validate the real-time monitoring data using a multi-source data verification mechanism to obtain radiation data;

[0015] The fusion storage unit is used to write the radiation data into a distributed storage network based on the alliance chain in blocks, and use smart contracts to automatically control access to and log data. Then, big data technology is used to build an index structure that supports time, location, and radiation level for storage query.

[0016] Optionally, the real-time radiation monitoring module includes:

[0017] The equipment monitoring unit is equipped with smart wearable devices to collect medical staff's operational behavior data in real time. The operating table is equipped with an RFID reader and a touch screen to read the dosage data in real time.

[0018] an anomaly identification unit, configured to deploy a lightweight deep learning model on the edge computing unit, utilize the lightweight deep learning model to identify injection operation data, metering operation data, and packaging operation data in the operation behavior data and the dosage data, obtain an action event log, and then utilize an association rule algorithm and time series analysis to identify anomalies in the action event log, determine whether the operation behavior data deviates from the standard process, and if so, automatically record it as an abnormal operation behavior;

[0019] The risk assessment unit is used to integrate the operational behavior data and the radiation data to obtain a fused data set, perform feature extraction and quantification on the fused data set to obtain a multidimensional feature vector, and then perform risk identification and risk scoring based on the multidimensional feature vector to obtain a risk assessment result.

[0020] Optionally, the equipment monitoring unit includes:

[0021] An intelligent sensing subunit is used to equip medical staff with intelligent wearable devices, use the intelligent wearable devices to collect operational behavior data, and use the built-in microprocessor of the intelligent wearable devices to preprocess the operational behavior data to obtain high-dimensional motion feature sequences and injection locations; the intelligent wearable devices include intelligent gloves and intelligent wristbands with integrated multimodal sensors;

[0022] The dose data synchronization subunit is used to equip the operating console with an RFID reader and a touch screen, use the RFID reader to read the dose data, and bind the dose data with the high-dimensional action feature sequence to complete the synchronization between dose reading and action.

[0023] Optionally, the risk assessment unit includes:

[0024] a data fusion subunit, configured to asynchronously match the operation behavior data and the radiation data using a time synchronization window to obtain a fused data set; wherein the operation behavior data includes an operation timestamp, operator, action type, dosage information, and injection location, and the radiation data includes a monitoring timestamp, sensor location, and radiation value;

[0025] a feature quantization subunit, configured to extract and quantify features from the fused dataset to obtain a multidimensional feature vector; the feature vector includes radiation values, drug dosage values, a one-hot encoding or embedding vector of the action type, a spatial encoding of the injection location, a sensor location encoding, and an operator identity encoding;

[0026] The risk identification and analysis subunit is used to mine the association between operations and risks based on the multi-dimensional feature vector to perform risk assessment.

[0027] Optionally, the risk identification and analysis subunit includes:

[0028] a feature clustering component, configured to cluster the multidimensional feature vectors using a density peak clustering algorithm to identify high radiation risk points, based on the high radiation risk points;

[0029] The risk association component is used to use the Apriori algorithm to mine the association between operations and risks, and to perform risk scoring based on the association between the high radiation risk points and the operations and risks. The calculation expression of the risk score is:

[0030]

[0031] Among them, R k is the risk score of the k-th operation point, v k is the radiation dose value, v max The highest radiation dose in history, d k is the drug dose, d max is the maximum drug dose in history, S(α k ,p k ) is the risk weight function of action-injection position, α, β, and γ are weight coefficients;

[0032] The threshold determination component is used to preset a risk score threshold. When the risk score is greater than the risk score threshold, it is recorded as a high-risk operation point and a risk assessment result is output.

[0033] Optionally, the intelligent prediction and evaluation module includes:

[0034] a risk prediction unit, configured to combine the multidimensional feature vector with the risk score to obtain a training data set, and train a random forest model using the training data set to obtain an initial prediction model;

[0035] A risk probability prediction unit is configured to introduce a predefined risk probability prediction function based on the initial prediction model to obtain a risk prediction model, and use the risk prediction model to perform risk prediction on the real-time multi-dimensional feature vector to obtain a predicted risk level;

[0036] The calculation expression of the predicted risk level is:

[0037]

[0038] in, To predict the risk level, x is the input feature vector, T is the total number of trees in the forest, and h t (x) is the risk category probability prediction of the t-th tree for the input x, w t is the weight of the t-th tree, and C is the total number of risk level categories.

[0039] Optionally, the operation safety assessment module includes:

[0040] a standard rule unit, configured to structure the fused data set and semiconductor radiation sensor status data to obtain a standard operation record table and construct a safety standard rule base; the safety standard rule base includes dose limit rules, operation time rules, device status rules, and radiation threshold rules;

[0041] The security assessment unit is used to develop a security assessment engine, match and judge the standard operation record table with the security standard rule base, obtain violation results, perform security scoring based on the violation results, and generate a security assessment report based on the security score; the calculation expression of the security score is:

[0042]

[0043] Among them, S is the security score, N is the total number of operations, M is the number of rule types, and w m is the weight of the mth rule, I k,m It is the judgment result of whether the k-th operation violates the m-th rule.

[0044] Optionally, the diversified intelligent warning module includes:

[0045] A comprehensive judgment unit is used to combine the safety score, the risk score and the predicted risk level according to the weight to obtain a comprehensive score, and then obtain the comprehensive score risk level according to a preset comprehensive score threshold; the calculation expression of the comprehensive score is:

[0046]

[0047] Among them, C is the comprehensive score, wS is the safety score weight, w R is the risk score weight, w P is the predicted risk level weight, S is the safety score, R is the risk score, P is the predicted risk level, ε is a small constant to prevent division by zero, and λ is the nonlinear adjustment coefficient;

[0048] The alarm unit is used to perform graded alarms and early warnings based on the comprehensive score.

[0049] Optionally, the alarm unit includes:

[0050] The alarm trigger subunit is used to generate graded alarms based on the comprehensive risk score through alarm sounds of different frequencies and rhythms and warning lights of different colors, and push the alarm information to the medical staff's mobile phone and the touch screen of the operation console in real time, and display the alarm information in a pop-up window;

[0051] The early warning auxiliary subunit is used to calculate the comprehensive score change rate based on the short-term change curve of the comprehensive score and preset a change rate threshold. When the comprehensive score change rate exceeds the change rate threshold, a risk early warning is triggered to remind medical staff of potential risks.

[0052] The present invention provides a radioactive operation safety assessment and warning system for nuclear medicine medication nursing, which discloses the following technical effects:

[0053] 1. Highly real-time radiation monitoring: 1) By deploying semiconductor radiation sensors within the nuclear medicine medication operation area, a spatial monitoring network is formed to achieve real-time monitoring of changes in radioactive materials. This cross-validation of monitoring data is performed by combining radiation data, operational behavior, and environmental sensor data, ensuring data integrity and quality, greatly reducing the risk of data errors and loss, and providing a reliable data foundation for subsequent risk analysis. 2) Through smart wearable devices and RFID readers, motion trajectories and doses can be collected synchronously in real time. Through edge computing nodes, abnormal operational behavior can be detected and identified in real time, while reducing data transmission pressure and improving processing speed and response time.

[0054] 2. Intelligent Prediction and Precise Quantification of Risk Assessment: 1) By combining density peak clustering and association rule mining, high-risk data clusters are identified, revealing patterns in the correlation between actions and radiation exposure. This allows for in-depth correlation analysis between radiation values and operational data, enabling better quantitative risk scoring. 2) By constructing a multidimensional risk prediction model based on fused features, scientific predictions of potentially high-risk operations can be made, enabling refined differentiation of risk levels and accurate prediction of radiation exposure probability, enhancing the foresight of risk response.

[0055] 3. Full-Process Operational Safety Assessment: 1) The safety assessment engine automatically matches and evaluates standard operation records against the safety standard rule base, enabling standardized and automated review of operational processes and reducing human judgment errors. 2) The standard operation record covers operational information and sensor device status throughout the entire process, enabling accurate and quantifiable operational safety scores, further improving operational safety levels and strengthening the standardization and scientific nature of operational process management.

[0056] 4. Diversified Risk Alerts: 1) A weighted comprehensive score is calculated for safety, risk, and predicted risk levels, dynamically assessing risk levels and providing differentiated alerts with varying tones and lighting colors. Simultaneously, real-time information is pushed through mobile devices and the console's touchscreen, enabling instant interaction and feedback. This enables multi-channel, multi-sensory interactive alerts, effectively enhancing medical staff's risk awareness and response speed. 2) An early warning assistance mechanism, based on the changing trends of the comprehensive risk score, can dynamically trigger early warnings to further reduce operational risks.

[0057] The technical solution of the present invention is further described in detail below through the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0058] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0059] Figure 1 A schematic diagram of the system architecture provided by an embodiment of the present invention;

[0060] Figure 2 A schematic diagram of a risk assessment process according to an embodiment of the present invention;

[0061] Figure 3 A schematic diagram of a risk prediction process according to an embodiment of the present invention;

[0062] Figure 4 A schematic diagram of the process of operational security assessment provided by an embodiment of the present invention;

[0063] Figure 5 A schematic diagram of the process of providing diversified warnings according to an embodiment of the present invention. DETAILED DESCRIPTION

[0064] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0065] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0066] like Figure 1 As shown, the present invention provides a radioactive operation safety assessment and warning system for nuclear medicine medication nursing, including interconnected data acquisition and processing modules, real-time radiation monitoring modules, intelligent prediction and assessment modules, operation safety assessment modules, and diversified intelligent warning modules.

[0067] 1. Data acquisition and processing module

[0068] It is used to deploy a space monitoring network in the nuclear medicine drug operation area, use the space monitoring network to monitor the changes of radioactive materials in real time, obtain radiation data, and use a distributed storage network to store and manage the radiation data; the data acquisition and processing module includes:

[0069] 1.1 Radiation sensor unit

[0070] It is used to deploy semiconductor radiation sensors, such as silicon photomultiplier tubes or high-purity germanium detectors, in the nuclear medicine operation area to ensure the detection of changes in low-dose radioactive substances, and design a sensor automatic calibration mechanism in the semiconductor radiation sensor to complete the deployment of the operation space monitoring network.

[0071] The nuclear medicine drug operation area includes the drug preparation area, injection area and waste disposal area. Multiple sensors are scientifically arranged to form a multi-point spatial monitoring network to achieve full coverage and redundant monitoring.

[0072] 1.2 Data initial processing unit

[0073] It is used to transmit the monitoring data of the semiconductor radiation sensor to the edge computing unit in real time, perform preliminary processing and data integration of the monitoring data, obtain a real-time monitoring data set, and then transmit the real-time monitoring data set to the cloud central control system for processing.

[0074] Preliminary processing, such as filtering and noise reduction, extracting key features such as instantaneous radiation intensity and cumulative dose, preliminarily determines the rationality of the data and achieves rapid response.

[0075] 1.3 Data Cleaning Unit

[0076] It is used to perform data cleaning, missing value filling, outlier extraction and time series synchronization operations on the real-time monitoring data set based on the cloud-based central control system to ensure data quality. It is then combined with operational behavior data and environmental sensor data, and a multi-source data verification mechanism is used to cross-validate the real-time monitoring data to further improve data accuracy and obtain radiation data.

[0077] 1.4 Fusion Storage Unit

[0078] This system writes radiation data, in blocks, to a distributed storage network based on a consortium blockchain. Each block contains a timestamp, sensor ID, radiation value, and verification information. Smart contracts are used to automatically control access to and log data, ensuring data security, privacy protection, and legal and compliant use. Big data technology is then used to construct an index structure that supports time, location, and radiation level, enabling rapid storage and query. This allows security managers and intelligent algorithms to access monitoring data in real time for further analysis and alert assessment.

[0079] 2. Real-time radiation monitoring module

[0080] like Figure 2 As shown, it is used to use smart wearable devices to collect medical staff's operation behavior data in real time, and deploy RFID readers and touch screens on the operating table to read dosage data in real time. Then, based on the operation behavior data and the dosage data, abnormality identification is performed to obtain abnormal operation behavior. Based on the operation behavior data and the radiation data, risk scoring is performed to obtain high-risk operation points and complete risk assessment. The real-time radiation monitoring module includes:

[0081] 2.1 Equipment Monitoring Unit

[0082] It is used to be equipped with smart wearable devices to collect the operation behavior data of medical staff in real time, and to be equipped with an RFID reader and a touch screen on the operating table to read the dosage data in real time; the equipment monitoring unit includes:

[0083] 2.1.1 Intelligent Perception Subunit

[0084] It is used to equip medical staff with smart wearable devices, use the smart wearable devices to collect operation behavior data, and use the built-in microprocessor of the smart wearable devices to preprocess the operation behavior data to obtain high-dimensional action feature sequences and injection positions.

[0085] The smart wearable device includes a smart glove and a smart wristband integrated with multimodal sensors; the multimodal sensors, such as accelerometers, gyroscopes, near-field communications (NFC) or RFID, pressure sensors, etc., can capture data on hand motion trajectory, drug dosage, and injection location in real time.

[0086] 2.1.2 Dose Data Synchronization Subunit

[0087] It is used to equip an operating table with an RFID reader and a touch screen, use the RFID reader to read dosage data, and bind the dosage data with the high-dimensional action feature sequence to complete wireless synchronization between dosage reading and action.

[0088] 2.2 Abnormal Recognition Unit

[0089] It is used to deploy a lightweight deep learning model on the edge computing unit, use the lightweight deep learning model to identify the injection operation data, metering operation data and packaging operation data in the operation behavior data and the dosage data, obtain an action event log, and then use the association rule algorithm and time series analysis to identify anomalies in the action event log to determine whether the operation behavior data deviates from the standard process, such as abnormal dosage or abnormal injection position. If the judgment result is yes, it is automatically recorded as an abnormal operation behavior.

[0090] 2.3 Risk Assessment Unit

[0091] It is used to integrate the operation behavior data and the radiation data to obtain a fused data set, extract and quantify the features of the fused data set to obtain a multi-dimensional feature vector, and then perform risk identification and risk scoring based on the multi-dimensional feature vector to obtain a risk assessment result. The risk assessment unit includes:

[0092] 2.3.1 Data Fusion Subunit

[0093] Used to use a time synchronization window to asynchronously match the operation behavior data and the radiation data to obtain a fused data set; wherein the operation behavior data includes an operation timestamp, operator, action type, dosage information and injection position, and the radiation data includes a monitoring timestamp, sensor position and radiation value.

[0094] 2.3.2 Feature Quantization Subunit

[0095] Used to extract and quantify the features in the fused data set to obtain a multidimensional feature vector; the feature vector includes radiation value, drug dose value, one-hot encoding or embedding vector of action type, spatial encoding of injection position, sensor position encoding and operator identity encoding.

[0096] 2.3.3 Risk Identification and Analysis Subunit

[0097] It is used to mine the association between operation and risk based on the multi-dimensional feature vector to perform risk assessment. The risk identification and analysis subunit includes:

[0098] The feature clustering component is used to cluster the multidimensional feature vectors using a density peak clustering algorithm to identify high radiation risk points.

[0099] The risk association component uses the Apriori algorithm to mine the association between operations and risks based on the high radiation risk points, and performs risk scoring based on the association between the high radiation risk points and operations and risks. The calculation expression of the risk score is:

[0100]

[0101] Among them, R k is the risk score of the k-th operation point, ranging from [0,1], v k is the radiation dose value, v max The highest radiation dose in history, d k is the drug dose, d max is the maximum drug dose in history, S(α k ,p k )∈[0,1] is the risk weight function of the action-injection position, and α, β, and γ are all weight coefficients.

[0102] The threshold determination component is used to preset a risk score threshold. When the risk score is greater than the risk score threshold, it is recorded as a high-risk operation point and a risk assessment result is output.

[0103] 3. Intelligent prediction and evaluation module

[0104] like Figure 3 As shown, it is used to build a risk prediction model, use the risk prediction model to predict the risk of real-time data, and obtain the predicted risk level; the intelligent prediction and evaluation module includes:

[0105] 3.1 Risk Prediction Unit

[0106] It is used to combine the multidimensional feature vector with the risk score to obtain a training data set, and use the training data set to train a random forest model to obtain an initial prediction model.

[0107] 3.2 Risk Probability Prediction Unit

[0108] It is used to introduce a predefined risk probability prediction function based on the initial prediction model to obtain a risk prediction model, and use the risk prediction model to perform risk prediction on the real-time multi-dimensional feature vector to obtain a predicted risk level.

[0109] The calculation expression of the predicted risk level is:

[0110]

[0111] in, To predict the risk level, x is the input feature vector, T is the total number of trees in the forest, and h t (x) is the risk category probability prediction of the t-th tree for the input x, w t is the weight of the t-th tree, and C is the total number of risk level categories; is a predefined risk probability prediction function.

[0112] 4. Operational safety assessment module

[0113] like Figure 4 As shown, it is used to build a security standard rule base, based on which a preset security assessment engine is used to automatically calculate the security score of an operation and generate a security assessment report; the operation security assessment module includes:

[0114] 4.1 Standard Rule Unit

[0115] It is used to structure the fused data set and semiconductor radiation sensor status data, obtain a standard operation record table, and build a safety standard rule base.

[0116] The standard operation record sheet includes: drug dosage, operation action, injection location, radiation value, operation time, equipment calibration status, operator identity and qualification information.

[0117] The safety standard rule base includes dose limitation rules, operation time rules, equipment status rules and radiation threshold rules.

[0118] 4.2 Safety Assessment Unit

[0119] Used to develop a security assessment engine, match and judge the standard operation record table with the security standard rule base, obtain violation results, perform security scoring based on the violation results, and generate a security assessment report based on the security score; the calculation expression of the security score is:

[0120]

[0121] Among them, S∈[0,1] is the security score. The closer the value is to 1, the safer the operation is. N is the total number of operations, M is the number of rule types, and w m is the weight of the mth rule, I k,m ∈{0,1} is the judgment result of whether the k-th operation violates the m-th rule, violation is 1 and compliance is 0.

[0122] The security assessment engine process is as follows:

[0123] 1) Read each piece of data in the operation record table;

[0124] 2) For each piece of data, match security rules one by one to determine compliance;

[0125] 3) Classify and attribute detected violations, such as abnormal dosage, excessive time, equipment abnormality, etc.

[0126] Design a graded violation marking system that assigns a grade [0, 1, 2, 3] based on the severity of the violation, where 0 is compliance and 3 is a severe violation.

[0127] The safety assessment report includes: operational compliance rate, distribution and detailed description of violation types, historical trends of key indicators such as time, dosage, equipment status, improvement suggestions and early warning measures.

[0128] 5. Diversified intelligent warning modules

[0129] like Figure 5 As shown, it is used to integrate the safety score, the risk score and the predicted risk level into a comprehensive score, and then perform graded sound and light alarms based on the comprehensive score; the diversified intelligent warning module includes:

[0130] 5.1 Comprehensive Judgment Unit

[0131] It is used to combine the safety score, the risk score and the predicted risk level according to the weight to obtain a comprehensive score, and then obtain the comprehensive score risk level according to the preset comprehensive score threshold; the calculation expression of the comprehensive score is:

[0132]

[0133] Among them, C is the comprehensive score. The higher the risk, the smaller the value. S is the safety score weight, w R is the risk score weight, w P is the predicted risk level weight, S is the safety score, R is the risk score, P is the predicted risk level, ε is a small constant to prevent division by zero, and λ>0 is the nonlinear adjustment coefficient to improve the response sensitivity to the highest risk item.

[0134] Determine the risk level based on the preset thresholds {T1, T2, T3}:

[0135] C≥T1: low risk level;

[0136] T2≤C<T1: medium risk level;

[0137] T3≤C<T2: high risk level;

[0138] C<T3: extremely high risk level.

[0139] 5.2 Alarm unit, used to issue graded alarms and early warnings based on the comprehensive score. The alarm unit includes:

[0140] Alarm trigger subunit

[0141] It is used to issue graded alarms based on the comprehensive risk score through alarm sounds of different frequencies and rhythms and warning lights of different colors, and push alarm information in real time to medical staff's mobile phones and the touch screen of the operation console, and display the alarm information in a pop-up window; the graded alarm method is shown in Table 1 below:

[0142] Table 1 Classification of alarms

[0143]

[0144] 5.3 Early Warning Auxiliary Subunit

[0145] It is used to calculate the comprehensive score change rate based on the short-term change curve of the comprehensive score and preset a change rate threshold. When the comprehensive score change rate exceeds the change rate threshold, a risk warning is triggered to remind medical staff of potential risks and ensure timely response.

[0146] Therefore, the present invention provides a radioactive operation safety assessment and warning system for nuclear medicine medication nursing. Through efficient and real-time data monitoring and intelligent and accurate risk assessment and operation safety assessment, it can effectively reduce the risk of radioactive exposure in nuclear medicine medication nursing, ensure the safety of medical staff and patients, and improve the scientificity and standardization of medical operations.

[0147] The various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the various embodiments can be referenced to each other.

[0148] This document uses specific examples to illustrate the principles and implementation methods of the present invention. The above examples are only intended to help understand the method and core concept of the present invention. At the same time, those skilled in the art will find that the specific implementation methods and application scopes may vary based on the concept of the present invention. In summary, the contents of this specification should not be construed as limiting the present invention.

Claims

1. A radioactive operation safety assessment and warning system for nuclear medicine medication nursing, characterized by: include: The data acquisition and processing module is used to deploy a space monitoring network in the nuclear medicine medication operation area, use the space monitoring network to monitor the changes of radioactive materials in real time, obtain radiation data, and use a distributed storage network to store and manage the radiation data; The real-time radiation monitoring module is used to collect the operation behavior data of medical staff in real time using smart wearable devices, and deploy RFID readers and touch screens on the operating table to read the dose data in real time. Based on the operation behavior data and the dose data, anomalies are identified to obtain abnormal operation behaviors. Based on the operation behavior data and the radiation data, risk scoring is performed to obtain high-risk operation points and complete risk assessment; An intelligent prediction and assessment module is used to build a risk prediction model, use the risk prediction model to predict the risk of real-time data, and obtain the predicted risk level; An operation safety assessment module is used to build a safety standard rule base, automatically calculate the safety score of the operation based on the safety standard rule base and use a preset safety assessment engine to generate a safety assessment report; A diversified intelligent warning module is used to integrate the safety score, the risk score and the predicted risk level into a comprehensive score, and then perform graded sound and light alarms based on the comprehensive score; Among them, the data acquisition and processing module, the real-time radiation monitoring module, the intelligent prediction and evaluation module, the operation safety evaluation module, and the diversified intelligent warning module are interconnected.

2. A radioactive operation safety assessment and warning system for nuclear medicine medication nursing according to claim 1, characterized in that: The data acquisition and processing module includes: Radiation sensor units are used to deploy semiconductor radiation sensors within the nuclear medicine administration area to detect changes in radioactive substances, and to design an automatic sensor calibration mechanism within the semiconductor radiation sensors to complete the deployment of the operation space monitoring network; A data initial processing unit is used to transmit the monitoring data of the semiconductor radiation sensor to the edge computing unit in real time, perform preliminary processing and data integration of the monitoring data to obtain a real-time monitoring data set, and then transmit the real-time monitoring data set to the cloud central control system for processing; A data cleaning unit is used to perform data cleaning, missing value filling, outlier extraction and time series synchronization operations on the real-time monitoring data set based on the cloud central control system, and then combine the operation behavior data and environmental sensor data to cross-validate the real-time monitoring data using a multi-source data verification mechanism to obtain radiation data; The fusion storage unit is used to write the radiation data into a distributed storage network based on the alliance chain in blocks, and use smart contracts to automatically control access to and log data. Then, big data technology is used to build an index structure that supports time, location, and radiation level for storage query.

3. A radioactive operation safety assessment and warning system for nuclear medicine medication nursing according to claim 2, characterized in that: The real-time radiation monitoring module includes: The equipment monitoring unit is equipped with smart wearable devices to collect medical staff's operational behavior data in real time. The operating table is equipped with an RFID reader and a touch screen to read the dosage data in real time. an anomaly identification unit, configured to deploy a lightweight deep learning model on the edge computing unit, utilize the lightweight deep learning model to identify injection operation data, metering operation data, and packaging operation data in the operation behavior data and the dosage data, obtain an action event log, and then utilize an association rule algorithm and time series analysis to identify anomalies in the action event log, determine whether the operation behavior data deviates from the standard process, and if so, automatically record it as an abnormal operation behavior; The risk assessment unit is used to integrate the operational behavior data and the radiation data to obtain a fused data set, perform feature extraction and quantification on the fused data set to obtain a multidimensional feature vector, and then perform risk identification and risk scoring based on the multidimensional feature vector to obtain a risk assessment result.

4. A radioactive operation safety assessment and warning system for nuclear medicine medication nursing according to claim 3, characterized in that: The equipment monitoring unit includes: An intelligent sensing subunit is used to equip medical staff with intelligent wearable devices, use the intelligent wearable devices to collect operational behavior data, and use the built-in microprocessor of the intelligent wearable devices to preprocess the operational behavior data to obtain high-dimensional motion feature sequences and injection locations; the intelligent wearable devices include intelligent gloves and intelligent wristbands with integrated multimodal sensors; The dose data synchronization subunit is used to equip the operating console with an RFID reader and a touch screen, use the RFID reader to read the dose data, and bind the dose data with the high-dimensional action feature sequence to complete the synchronization between dose reading and action.

5. A radioactive operation safety assessment and warning system for nuclear medicine medication nursing according to claim 4, characterized in that: The risk assessment unit includes: a data fusion subunit, configured to asynchronously match the operation behavior data and the radiation data using a time synchronization window to obtain a fused data set; wherein the operation behavior data includes an operation timestamp, operator, action type, dosage information, and injection location, and the radiation data includes a monitoring timestamp, sensor location, and radiation value; a feature quantization subunit, configured to extract and quantify features from the fused dataset to obtain a multidimensional feature vector; the feature vector includes radiation values, drug dosage values, a one-hot encoding or embedding vector of the action type, a spatial encoding of the injection location, a sensor location encoding, and an operator identity encoding; The risk identification and analysis subunit is used to mine the association between operations and risks based on the multi-dimensional feature vector to perform risk assessment.

6. A radioactive operation safety assessment and warning system for nuclear medicine medication nursing according to claim 5, characterized in that: The risk identification and analysis subunit includes: a feature clustering component, configured to cluster the multidimensional feature vectors using a density peak clustering algorithm to identify high radiation risk points, based on the high radiation risk points; The risk association component is used to use the Apriori algorithm to mine the association between operations and risks, and to perform risk scoring based on the association between the high radiation risk points and the operations and risks. The calculation expression of the risk score is: Among them, R k is the risk score of the k-th operation point, v k is the radiation dose value, v max The highest radiation dose in history, d k is the drug dose, d max is the maximum drug dose in history, S(α k ,p k ) is the risk weight function of action-injection position, α, β, and γ are weight coefficients; The threshold determination component is used to preset a risk score threshold. When the risk score is greater than the risk score threshold, it is recorded as a high-risk operation point and a risk assessment result is output.

7. A radioactive operation safety assessment and warning system for nuclear medicine medication nursing according to claim 6, characterized in that: The intelligent prediction and evaluation module includes: a risk prediction unit, configured to combine the multidimensional feature vector with the risk score to obtain a training data set, and train a random forest model using the training data set to obtain an initial prediction model; A risk probability prediction unit is configured to introduce a predefined risk probability prediction function based on the initial prediction model to obtain a risk prediction model, and use the risk prediction model to perform risk prediction on the real-time multi-dimensional feature vector to obtain a predicted risk level; The calculation expression of the predicted risk level is: in, To predict the risk level, x is the input feature vector, T is the total number of trees in the forest, and h t (x) is the risk category probability prediction of the t-th tree for the input x, w t is the weight of the t-th tree, and C is the total number of risk level categories.

8. A radioactive operation safety assessment and warning system for nuclear medicine medication nursing according to claim 7, characterized in that: The operational safety assessment module includes: a standard rule unit, configured to structure the fused data set and semiconductor radiation sensor status data to obtain a standard operation record table and construct a safety standard rule base; the safety standard rule base includes dose limit rules, operation time rules, device status rules, and radiation threshold rules; The security assessment unit is used to develop a security assessment engine, match and judge the standard operation record table with the security standard rule base, obtain violation results, perform security scoring based on the violation results, and generate a security assessment report based on the security score; the calculation expression of the security score is: Among them, S is the security score, N is the total number of operations, M is the number of rule types, and w m is the weight of the mth rule, I k,m It is the judgment result of whether the k-th operation violates the m-th rule.

9. A radioactive operation safety assessment and warning system for nuclear medicine medication nursing according to claim 8, characterized in that: The diversified intelligent warning module includes: A comprehensive judgment unit is used to combine the safety score, the risk score and the predicted risk level according to the weight to obtain a comprehensive score, and then obtain the comprehensive score risk level according to a preset comprehensive score threshold; the calculation expression of the comprehensive score is: Among them, C is the comprehensive score, w S is the safety score weight, w R is the risk score weight, w P is the predicted risk level weight, S is the safety score, R is the risk score, P is the predicted risk level, ε is a small constant to prevent division by zero, and λ is the nonlinear adjustment coefficient; The alarm unit is used to perform graded alarms and early warnings based on the comprehensive score.

10. A radioactive operation safety assessment and warning system for nuclear medicine medication nursing according to claim 9, characterized in that: The alarm unit comprises: The alarm trigger subunit is used to generate graded alarms based on the comprehensive risk score through alarm sounds of different frequencies and rhythms and warning lights of different colors, and push the alarm information to the medical staff's mobile phone and the touch screen of the operation console in real time, and display the alarm information in a pop-up window; The early warning auxiliary subunit is used to calculate the comprehensive score change rate based on the short-term change curve of the comprehensive score and preset a change rate threshold. When the comprehensive score change rate exceeds the change rate threshold, a risk early warning is triggered to remind medical staff of potential risks.

Images