Emergency rescue room medicine intelligent management and accurate distribution system and management method

Abstract

The invention discloses an emergency rescue room medicine intelligent management and accurate distribution system and management method, and relates to the technical field of medical informationization, the system comprises an information acquisition module, an intelligent medicine use decision module, a medicine sorting and distribution execution module, a medicine use monitoring and feedback module and a system optimization iteration module; by constructing the drug interaction rule base and the risk assessment model, intelligent inspection and quantitative assessment of incompatibility are achieved, the rule base covers chemical incompatibility, basic incompatibility of metabolic pathway conflicts and patient specific incompatibility associated with allergy history and basic diseases, and the risk assessment model has the advantages that the risk assessment model and the risk assessment model are combined, so that the risk assessment of the incompatibility is achieved. A multi-dimensional risk prevention and control network is formed in combination with a time window of an emergency treatment scene and multi-channel medication taboo, a risk assessment model quantifies a medication risk value and sets a threshold value through dynamic weighted calculation of physical sign parameters, medical history risk factors and time attenuation coefficients, and when the risk exceeds the standard, early warning is automatically triggered.

Description

Technical Field

[0001] This invention relates to the field of medical information technology, specifically to an intelligent drug management and precise distribution system and management method for emergency resuscitation rooms. Background Technology

[0002] As a high-risk scenario in the healthcare system, the efficiency and accuracy of medication management in the emergency room directly impact patient prognosis. In the traditional model, doctors need to manually check drug incompatibilities, calculate dosages, and assess risks, while nurses rely on experience to sort and distribute medications. This process is highly dependent on manual operation, resulting in fragmented information and delayed response. With the development of medical informatization, although hospitals have deployed HIS and EMR systems, the data silos between these systems are severe, and patient dynamic signs and medication information have not been effectively integrated. In addition, the special nature of the emergency room further exacerbates medication risks. According to statistics, drug incompatibilities and dosage errors account for more than 60% of medication errors in the emergency room, and traditional monitoring methods are difficult to achieve full-process coverage. Therefore, developing an intelligent management system that can collect multi-source data in real time, automate decision-making, and perform closed-loop optimization has become a key requirement for improving the safety of emergency medication.

[0003] Traditional emergency medication management relies heavily on manual experience and static rules, resulting in three major drawbacks: First, insufficient data integration, with patient basic information, dynamic vital signs, and medication data scattered across different systems, requiring medical staff to search across platforms, which can easily lead to information omissions or misinterpretations. Second, low decision-making efficiency, as drug incompatibilities checks rely on pharmacists manually verifying clinical guidelines, which is time-consuming and susceptible to subjective factors; in emergency scenarios, the dynamic priority of patient urgency and medication time windows cannot be quantified, leading to an imbalance in resource allocation. Third, delayed monitoring and feedback, with traditional methods relying on post-hoc checks or passive reporting to detect medication abnormalities, making it difficult to intervene in high-risk patients in real time. Furthermore, traditional systems lack an iterative mechanism, with contraindications and model parameter updates relying on manual maintenance, failing to adapt to rapid changes in clinical needs. Summary of the Invention

[0004] The purpose of this invention is to overcome the shortcomings of existing technologies and provide an intelligent drug management and precise distribution system and method for emergency resuscitation rooms. By integrating multi-source data from hospitals, a drug interaction rule base and risk assessment model are constructed to achieve intelligent checking and quantitative assessment of drug incompatibilities. The system utilizes a dynamic priority model for emergency scenarios to optimize drug sorting and early warning sequences, and combines intelligent medicine cabinets and AGV robots to complete automated distribution, significantly shortening medication response time. Simultaneously, the medication monitoring and feedback module supports tiered early warning and process backtracking, forming a data-driven closed-loop optimization mechanism.

[0005] To solve the above-mentioned technical problems, the present invention provides the following technical solution: On the one hand, an intelligent drug management and precise distribution system for emergency resuscitation rooms, the system comprising: an information acquisition module, an intelligent medication decision-making module, a drug sorting and distribution execution module, a medication monitoring and feedback module, and a system optimization and iteration module; The information collection module connects to the hospital's HIS and EMR systems to collect basic patient information, uses multi-parameter monitors and trauma assessment terminals to obtain dynamic vital signs and medication time windows, and performs de-identification processing. At the same time, it builds a drug information database, loads drug instructions, clinical guidelines, and inventory data, and preprocesses and standardizes the data. The intelligent medication decision-making module: When a doctor writes a prescription or a nurse picks up medication, it calls the drug interaction rule base and the drug use risk assessment model to check for drug incompatibilities. It assesses the medication risk and outputs intervention suggestions through a drug incompatibilities quantification model that integrates patient characteristics. Based on the patient's level of urgency and medication time window, it uses a dynamic priority model for emergency scenarios to calculate drug sorting and early warning priorities. The drug sorting and distribution execution module: after receiving intelligent decision instructions, controls the intelligent medicine cabinet and automatic sorting machine to complete drug sorting, and then distributes the drugs through the transmission track and AGV robot, and records the data of the entire sorting and distribution process; The medication monitoring and feedback module connects to the monitoring system and LIS system to collect vital signs and laboratory data of high-risk patients. Combined with subjective reaction data, it uses an anomaly recognition model to identify medication abnormalities and trigger graded early warnings. At the same time, it traces back the medication process to generate reports and supports medical staff annotations and adjustments. The system optimization and iteration module updates the drug information database daily, adds new contraindication cases and adjusts inventory, calls up full-process data weekly, optimizes model parameters using parameter optimization algorithms, and verifies and updates the system to the production environment after iteration.

[0006] Furthermore, the drug interaction rule base in the intelligent medication decision-making module contains the following content: Basic drug combination contraindications: chemical incompatibility, drug antagonistic combinations, combinations with superimposed toxic and adverse reactions, and combinations with conflicting metabolic pathways; Patient-specific drug interaction contraindications: These include contraindications related to a history of allergies and contraindications related to underlying diseases; Special contraindications for medication in emergency settings: contraindications related to time windows and contraindications for medication administered through multiple channels.

[0007] Furthermore, in the intelligent medication decision-making module, the calculation formula for constructing the drug use risk assessment model is as follows: ,in, It is a drug use risk assessment value, used to quantify the risk of drug use. It is an index of the patient's vital signs parameters, from 1 to... , representing different vital signs indicators, It is the total number of the patient's vital signs parameters. It is the weight of the patient's vital signs. These are normalized values ​​of vital signs. No. The raw values ​​of the patient's vital signs, The patient's medical history is a "risk signal": allergy history = 1, no allergy = 0. No. Patient medical history information, It is an index of patient medical history and risk factors, from 1 to... , indicating different types of medical history It is the total number of risk factors in the patient's medical history. It is the correlation between medical history and current medications. This is information about the drugs currently in use. It is the time decay coefficient. It is the time difference between the current time and the time of the most recent valid vital sign data collection, combined with the drug interaction rule base to set the drug use risk assessment threshold. ,when An alert will be triggered if the threshold is exceeded.

[0008] Furthermore, the calculation formula for constructing the drug incompatibility quantification model in the intelligent medication decision-making module is as follows: , This is a quantitative value for drug incompatibilities, used to quantify the degree of contraindication when two drugs are used together. This involves information on two drugs whose drug incompatibilities are to be evaluated. It is based on graph neural network (GNN) computation. and The interaction values ​​between two drugs reflect the risk of drug incompatibility. Graph Neural Networks (GNNs) employ a multi-relationship heterogeneous graph structure, where nodes represent drug chemical components / targets and edges represent interaction types. Embedding vectors are obtained through pre-training on a pharmaceutical database. It is an index of individual patient characteristic parameters, from 1 to... , representing different individual characteristics It is the total number of individual patient characteristic parameters. No. Patient individual characteristics information, No. Individual characteristics of patients and Factors influencing drug incompatibilities reflect the impact of individual differences on incompatibilities, and quantitative thresholds for drug incompatibilities are set based on clinical data. ,when Trigger an early warning and provide intervention suggestions.

[0009] Furthermore, the specific content of the intervention suggestions in the intelligent medication decision-making module is as follows: prioritize recommending equivalent alternative drugs with different mechanisms of action; when substitution is not possible, adjust the drug dosage, interval or route of administration to reduce risk; clarify high-frequency monitoring indicators and emergency drugs and procedures; and provide personalized matching suggestions based on individual patient differences.

[0010] Furthermore, in the intelligent medication decision-making module, the calculation formula for constructing the dynamic priority model for emergency scenarios is as follows: ,in, This is a dynamic priority value for emergency scenarios, used to determine the priority of drug sorting and early warning. It contains patient information, including real-time vital signs data. It is based on real-time vital sign data from the patient information collection module. M represents drug information, including the drug's time window requirements. TimeWindow(M) is determined by the drug time window requirements in the drug information database, combined with the remaining time until the appropriate time to administer the medication. It is an index of sudden emergencies, from 1 to... This indicates different types of sudden emergencies. It is the total number of sudden emergencies. It is the first Emergency information When the system detects a sudden emergency, it assigns a value according to preset rules. Sort by numerical value The higher the value, the higher the priority of drug sorting and early warning.

[0011] Furthermore, in the medication monitoring and feedback module, the anomaly identification model is calculated using the following formula: ,in, These are medication anomaly detection values ​​used to determine whether a patient's condition becomes abnormal after medication administration. It refers to the time point after medication is administered. These are changes in physical signs after medication. This is the baseline of physical signs before medication. This is an expected change in the patient's condition. This is the baseline expectation of the condition. It is a risk assessment value for drug use.

[0012] Furthermore, the classification of abnormal levels in the medication monitoring and feedback module is as follows: when When this occurs, a Level 1 warning is triggered, reminding medical staff to pay attention to the patient and continuously monitor changes in vital signs, without requiring proactive adjustments to medication. when When this is triggered, a level-two alert is issued, reminding medical staff to quickly check the patient's condition, review the medication regimen, and assess whether to adjust the dosage or change the medication. when When an emergency alert is triggered, medical staff immediately rush to the patient's side, initiate emergency response procedures, and simultaneously trace back the entire medication process to investigate the root cause.

[0013] Furthermore, in the system optimization iteration module, the model parameters are optimized using a parameter optimization algorithm, and the calculation formula is as follows: ,in, It is the optimized set of model parameters. It is the set of model parameters before optimization. It's the learning rate. These are the indexes of the sample data, from 1 to N, representing different sample data. It is the total number of sample data used for model parameter optimization. These are gradient values, representing the drug use risk assessment values ​​based on this sample. and actual results The calculations reflect the direction and magnitude of the model parameter adjustments. No. Drug use risk assessment value for each sample, No. The actual medication results of each sample It is a minimum value.

[0014] On the other hand, an intelligent medication management method for emergency resuscitation rooms includes the following steps: S100, Information Collection and Preprocessing: This system connects to the hospital's HIS and EMR systems to collect basic patient information. It utilizes multi-parameter monitors and trauma assessment terminals to collect dynamic vital signs and medication time windows, and performs de-identification processing. A drug information database is built and loaded with drug instructions, clinical guidelines, and inventory data. Simultaneously, the collected data undergoes preprocessing and standardization. S200, Intelligent Medication Decision-Making: When doctors prescribe medication or nurses collect medication, it calls up a drug interaction rule base that includes contraindications to basic drug combinations, patient-specific contraindications, and special contraindications for emergency scenarios. It calculates the risk value by combining the drug use risk assessment model, evaluates the degree of contraindication by using a drug compatibility quantification model, generates intervention suggestions, and calculates drug sorting and early warning priorities based on the patient's critical condition, medication time window, and sudden emergency events through a dynamic priority model for emergency scenarios. S300, Drug Sorting and Distribution Execution: Receives intelligent decision-making instructions, controls intelligent medicine cabinets and automatic sorting machines to complete drug sorting, handles drugs with special storage conditions as required, distributes drugs through transmission tracks or AGV robots, and records the entire process data of sorting time, drug information, distribution path, equipment status and operators. S400, Medication Monitoring and Feedback: The medication monitoring and feedback module connects to the monitoring system and LIS system to collect vital signs, laboratory data and subjective reaction data of high-risk patients. It uses an anomaly recognition model to calculate anomaly recognition values, classifies the warnings into basic, intermediate and emergency levels according to the value and triggers corresponding measures. At the same time, it traces back the medication process to generate reports and supports medical staff annotations and adjustments. S500, system optimization and iteration: daily updates to the drug information database, addition of new contraindication cases, adjustment of inventory, and updates to instructions and guidelines; weekly access to full-process data, optimization of model parameters using parameter optimization algorithms, iteration verification, and updates to the production environment.

[0015] Compared with existing technologies, this intelligent drug management and precise distribution system and management method for emergency resuscitation rooms has the following beneficial effects: I. This invention achieves intelligent inspection and quantitative assessment of drug incompatibilities by constructing a drug interaction rule base and risk assessment model. The rule base covers basic contraindications such as chemical incompatibilities and metabolic pathway conflicts, as well as patient-specific contraindications related to allergies and underlying diseases. Combined with the time window and multi-channel medication contraindications in emergency scenarios, a multi-dimensional risk prevention and control network is formed. The risk assessment model quantifies the medication risk value and sets a threshold through dynamic weighted calculation of vital signs, medical history risk factors, and time decay coefficients. When the risk exceeds the threshold, an early warning is automatically triggered. In addition, the drug incompatibility quantification model introduces a graph neural network to calculate the drug interaction value and adjusts the degree of contraindication based on individual patient characteristics, generating suggestions for equivalent alternative drugs and dosage adjustments. This precise decision-making mechanism reduces medication errors caused by human oversight.

[0016] Second, this invention utilizes a dynamic priority model for emergency scenarios to integrate real-time information on patient vital signs, medication time windows, and sudden emergencies. It dynamically calculates medication sorting and early warning priorities, automatically prioritizing medications for critically ill patients or those with urgent time windows. Automated distribution is achieved through intelligent medicine cabinets, AGV robots, and transport tracks, shortening medication response time. Simultaneously, the medication monitoring and feedback module connects to the monitoring system and LIS system, using an anomaly identification model to continuously monitor high-risk patients. Anomalies are categorized into Level 1, Level 2, and emergency warnings, generating medication process reports for medical staff to review. The system updates the medication database daily and iterates model parameters weekly, forming a data-driven continuous optimization loop. This end-to-end intelligent management not only improves emergency rescue efficiency but also reduces medical risks through a tiered early warning mechanism.

[0017] Other advantages, objectives and features of the invention will be set forth in part in the description which follows, and in part will be apparent to those skilled in the art from an examination of the following, or may be learned from the practice of the invention. Attached Figure Description

[0018] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the accompanying drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are merely some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without any creative effort.

[0019] Figure 1 This is a framework diagram of an intelligent drug management and precise distribution system for emergency resuscitation rooms. Figure 2 This is a flowchart of an intelligent drug management method for emergency resuscitation rooms. Detailed Implementation

[0020] To further illustrate the technical means and effects of the present invention in achieving its intended purpose, the following detailed description of the specific implementation methods, structures, features, and effects of the present invention, in conjunction with the accompanying drawings and preferred embodiments, is provided below.

[0021] Example 1: Scenario of medication management for patients with acute myocardial infarction.

[0022] Information Collection Module: When the emergency room receives a patient suffering from a sudden acute myocardial infarction, the system automatically connects to the hospital's HIS and EMR systems to quickly obtain basic information such as the patient's age, past medical history, and allergy history. Figure 1 As shown, the system simultaneously collects dynamic vital signs data such as heart rate, blood pressure, and blood oxygen saturation in real time through a multi-parameter monitor and performs de-identification processing. In addition, the system synchronously loads the instructions, clinical guidelines, and inventory data of aspirin and ticagrelor emergency drugs from the drug information database, and performs format unification and standardization processing on all collected data to provide complete and accurate data support for subsequent intelligent decision-making, ensuring the consistency and availability of information.

[0023] The intelligent medication decision-making module: When a doctor prescribes an antiplatelet aggregation medication containing aspirin and ticagrelor, the system immediately accesses the drug interaction rule base to comprehensively check for contraindications to the basic drug combination (such as chemical incompatibility and drug antagonism), patient-specific contraindications (such as a history of drug allergies), and special contraindications for emergency situations (such as whether the medication time window meets emergency requirements). Simultaneously, it utilizes a medication use risk assessment model to integrate patient vital signs and medical history data to quantify and calculate the medication risk value. The calculation formula is as follows: ,in, It is a drug use risk assessment value, used to quantify the risk of drug use. It is an index of the patient's vital signs parameters, from 1 to... , representing different vital signs indicators, It is the total number of the patient's vital signs parameters. It is the weight of the patient's vital signs. These are normalized values ​​of vital signs. No. The raw values ​​of the patient's vital signs, The patient's medical history is a "risk signal": allergy history = 1, no allergy = 0. No. Patient medical history information, It is an index of patient medical history and risk factors, from 1 to... , indicating different types of medical history It is the total number of risk factors in the patient's medical history. It is the correlation between medical history and current medications. This is information about the drugs currently in use. It is the time decay coefficient. It is the time difference between the current time and the time of the most recent valid vital sign data collection, combined with the drug interaction rule base to set the drug use risk assessment threshold. ,when If the threshold is exceeded, an early warning is triggered, and the degree of incompatibility between the two drugs is assessed using a drug incompatibility quantification model. The calculation formula is as follows: , This is a quantitative value for drug incompatibilities, used to quantify the degree of contraindication when two drugs are used together. This involves information on two drugs whose drug incompatibilities are to be evaluated. It is based on graph neural network (GNN) computation. and The interaction value between two drugs reflects the risk of incompatibility between the drugs themselves. It is an index of individual patient characteristic parameters, from 1 to... , representing different individual characteristics It is the total number of individual patient characteristic parameters. No. Patient individual characteristics information, No. Individual characteristics of patients and Factors influencing drug incompatibilities reflect the impact of individual differences on incompatibilities, and quantitative thresholds for drug incompatibilities are set based on clinical data. ,when The system triggers an alert and outputs intervention suggestions. These suggestions are automatically generated, such as recommending equivalent alternatives with different mechanisms of action, or adjusting dosage and administration route to reduce medication risks. Furthermore, the system assesses the severity of the emergency based on the patient's vital signs, considering the medication time window and the occurrence of sudden cardiac arrhythmias. It calculates drug sorting and alert priorities using a dynamic priority model for emergency scenarios, with the following formula: ,in, This is a dynamic priority value for emergency scenarios, used to determine the priority of drug sorting and early warning. It contains patient information, including real-time vital signs data. It is based on real-time vital sign data from the patient information collection module. M represents drug information, including the drug's time window requirements. TimeWindow(M) is determined by the drug time window requirements in the drug information database, combined with the remaining time until the appropriate time to administer the medication. It is an index of sudden emergencies, from 1 to... This indicates different types of sudden emergencies. It is the total number of sudden emergencies. It is the first Emergency information When the system detects a sudden emergency, it assigns a value according to preset rules. Sort by numerical value The higher the value, the higher the priority of drug sorting and early warning, ensuring that high-risk and emergency medications are handled first and optimizing the rescue process.

[0024] Drug sorting and distribution execution module: After receiving intelligent decision-making instructions, the system automatically controls the intelligent medicine cabinet and automatic sorting machine to accurately sort aspirin and ticagrelor from the inventory. For drugs that require special storage conditions such as protection from light and refrigeration, the system performs corresponding processing procedures to prevent drug deterioration. The drugs are quickly distributed to the emergency room via a transport track or AGV robot using the shortest path. The entire process records sorting time, drug batch, distribution path, equipment status, and operator data in real time, forming a complete traceability chain. This ensures both the timeliness of drug delivery and the controllability of quality throughout the entire process, providing hardware execution support for the accuracy and safety of emergency medication.

[0025] Medication Monitoring and Feedback Module: After medication administration, the system continuously connects to the monitoring system and LIS system to dynamically collect patient vital signs (such as heart rate fluctuations and blood pressure recovery), laboratory data (such as coagulation function indicators), and patient subjective reactions (such as the degree of chest pain relief). It uses an anomaly detection model to comprehensively analyze the difference between post-medication changes in vital signs and the expected course of illness, calculates anomaly detection values, and classifies warning levels. The calculation formula is as follows: ,in, These are medication anomaly detection values ​​used to determine whether a patient's condition becomes abnormal after medication administration. It refers to the time point after medication is administered. These are changes in physical signs after medication. This is the baseline of physical signs before medication. This is an expected change in the patient's condition. This is the baseline expectation of the condition. It is a drug use risk assessment value, when When this is triggered, a Level 1 alert is activated, reminding medical staff to continuously monitor the patient's condition; when At that time, a level-two warning is issued, prompting medical staff to quickly verify the medication regimen and the patient's response; In case of an emergency, an alert is issued, immediately notifying medical staff to initiate emergency response procedures. At the same time, the entire medication process is automatically traced back to generate a detailed report, supporting medical staff annotations and adjustments, enabling real-time monitoring and intervention of medication risks, and reducing the incidence of adverse reactions.

[0026] System optimization and iteration module: The drug information database is automatically updated daily, synchronously supplementing new cases of antiplatelet drug contraindications, adjusting inventory quantities, and updating the latest versions of drug instructions and clinical guidelines to ensure data timeliness. Weekly, the entire process data of the patient's resuscitation is retrieved, and parameter optimization algorithms are used to optimize the parameters of the drug use risk assessment model and the drug incompatibility quantification model. The calculation formula is as follows: ,in, It is the optimized set of model parameters. It is the set of model parameters before optimization. It's the learning rate. These are the indexes of the sample data, from 1 to N, representing different sample data. It is the total number of sample data used for model parameter optimization. These are gradient values, representing the drug use risk assessment values ​​based on this sample. and actual results The calculations reflect the direction and magnitude of the model parameter adjustments. No. Drug use risk assessment value for each sample, No. The actual medication results of each sample It is a minimum value. After the iterative model is clinically validated, it is updated to the production environment, enabling the system to continuously adapt to updates in medical guidelines and the needs of clinical practice, and continuously improve the intelligence and accuracy of drug management.

[0027] In summary, this intelligent drug management and precise distribution system for emergency resuscitation rooms integrates multi-source data through information collection and preprocessing in the resuscitation of patients with acute myocardial infarction. The intelligent medication decision-making module ensures medication safety with the help of various models and rule bases. Drug sorting and distribution execution achieves precise delivery and traceability. Medication monitoring and feedback promptly detect anomalies. System optimization and iteration continuously improve performance. The synergistic effect of each module forms a closed-loop management system from data collection to system optimization, providing accurate, safe, and efficient drug management and distribution services for patients with acute myocardial infarction, and helping to improve the quality and efficiency of emergency resuscitation.

[0028] Example 2: Emergency medication management for severely trauma patients.

[0029] S100, Information Collection and Preprocessing: An emergency room patient with multiple fractures and hemorrhagic shock due to a traffic accident is admitted. The system immediately connects to the hospital's HIS and EMR systems to obtain basic information such as the patient's gender, age, underlying medical history, and drug allergy history. Real-time monitoring of blood pressure, pulse, and respiratory rate is conducted using a multi-parameter monitor. The system accurately determines the golden time window for hemostasis and anti-shock treatment using a trauma assessment terminal and performs de-identification processing. Simultaneously, the system loads the instructions, clinical guidelines, and inventory data for tranexamic acid and ceftriaxone (hemostatic and anti-infective drugs) from the drug information database. All collected data is cleaned, formatted, and standardized to ensure accuracy and completeness, providing a reliable data foundation for subsequent drug management decisions and meeting the information needs of rapid emergency response. Figure 2 As shown.

[0030] S200, Intelligent Medication Decision-Making: When a nurse administers tranexamic acid and ceftriaxone to a patient, the system automatically calls upon a drug interaction rule base containing basic drug combination contraindications (such as whether there is a metabolic pathway conflict between tranexamic acid and ceftriaxone), patient-specific contraindications (such as whether the patient is allergic to cephalosporin antibiotics), and emergency-specific contraindications (such as whether there are incompatibility risks associated with multi-channel medication administration). The system then uses a drug use risk assessment model combined with the patient's trauma severity and vital signs data to quantify the medication risk. The calculation formula is as follows: In conjunction with a drug interaction rule base, set thresholds for drug use risk assessment. ,when If the threshold is exceeded, an early warning is triggered. A drug incompatibility quantification model is used to assess the degree of incompatibility between the two drugs used together. The calculation formula is as follows: Quantitative thresholds for drug incompatibilities set based on clinical data ,when The system triggers an alert and outputs intervention suggestions. If a high risk is detected, intervention suggestions are generated, such as changing the type of antibiotic or adjusting the dosing interval. Furthermore, the system calculates drug sorting and alert priorities based on the patient's trauma severity score, remaining medication time window, and whether a sudden hemorrhagic shock emergency has occurred, using a dynamic priority model for emergency scenarios. The calculation formula is as follows: ,right Sort by numerical value The higher the value, the higher the priority of drug sorting and early warning, ensuring that emergency medication is given priority, maximizing the timeliness requirements of trauma treatment, and reducing the risks to patients caused by medication delays.

[0031] S300, Drug Sorting and Distribution Execution: Based on decision-making instructions, the system controls intelligent medicine cabinets and automatic sorting machines to quickly locate and sort out tranexamic acid and ceftriaxone. For drugs requiring temperature storage, a constant temperature transport device is activated to prevent drug failure due to improper storage conditions. The drugs are distributed to the emergency room with maximum efficiency via a transport track or AGV robot. The system simultaneously records the sorting time, drug information, distribution equipment, and operator data throughout the entire process, forming a traceable electronic record. This not only ensures that drugs are delivered in the shortest possible time but also achieves full-process quality monitoring from drug issuance to use, providing technical support for the precise execution of trauma emergency medication.

[0032] S400, Medication Monitoring and Feedback: After medication administration, the system collects real-time data on the patient's vital signs (such as the rate of blood pressure recovery and heart rate stability), laboratory data (such as hemoglobin levels and inflammatory markers), and pain perception. An anomaly detection model analyzes the deviation between post-medication changes in vital signs and the expected disease progression, calculates anomaly detection values, and triggers corresponding warning levels. The calculation formula is as follows: ,when When this occurs, a Level 1 alert is triggered, and the patient's condition is continuously monitored; when When this is triggered, a level-two warning is issued, prompting medical staff to reassess the medication regimen; when When an emergency warning is triggered, the emergency response process is immediately initiated. At the same time, a report is generated by reviewing the entire medication process, allowing medical staff to analyze the root causes and adjust the treatment plan. This enables dynamic monitoring of trauma patients after medication, timely detection and handling of adverse drug reactions or poor efficacy, and ensures the safety of treatment.

[0033] S500, system optimization and iteration: The drug information database is automatically updated daily, adding new cases of drug contraindications related to trauma treatment, adjusting inventory strategies, and simultaneously updating the latest clinical guidelines and instructions for hemostatic drugs and antibiotics to ensure data is in sync with medical advancements. Weekly, the entire patient treatment process data is accessed, and parameter optimization algorithms are used to optimize the parameters of the drug use risk assessment model and anomaly identification model. The calculation formula is as follows: After the iterative model is verified through simulation and clinical testing, it is updated to the production environment, which continuously optimizes the drug management method, better meets the personalized treatment needs of severely trauma patients, and continuously improves the scientificity and effectiveness of emergency drug management.

[0034] In summary, this intelligent medication management method for emergency resuscitation rooms, in the emergency management scenario of severely trauma patients, acquires key data through information collection and preprocessing, generates reasonable suggestions and determines priorities by combining multiple models and rules for intelligent medication decision-making, ensures timely and accurate delivery of medications through medication sorting and distribution, safeguards medication safety through medication monitoring and feedback, and promotes method improvement through system optimization and iteration. The close integration of each step forms a scientific management process, enabling intelligent medication management and precise distribution tailored to the treatment characteristics of severely trauma patients, providing strong support for the emergency treatment of severely trauma patients, and improving the level of emergency medication management.

[0035] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention in any way. Although the present invention has been disclosed above with reference to preferred embodiments, it is not intended to limit the present invention. Any person skilled in the art can make some modifications or alterations to the above-disclosed technical content to create equivalent embodiments without departing from the scope of the present invention. Any simple modifications, equivalent changes and alterations made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the scope of the present invention.

Claims

1. An intelligent medication management and precise dispensing system for emergency resuscitation rooms, characterized in that, The system includes: an information acquisition module, an intelligent medication decision-making module, a drug sorting and distribution execution module, a medication monitoring and feedback module, and a system optimization and iteration module; The information collection module connects to the hospital's HIS and EMR systems to collect basic patient information, uses multi-parameter monitors and trauma assessment terminals to obtain dynamic vital signs and medication time windows, and performs de-identification processing. At the same time, it builds a drug information database, loads drug instructions, clinical guidelines, and inventory data, and preprocesses and standardizes the data. The intelligent medication decision-making module: When a doctor writes a prescription or a nurse picks up medication, it calls the drug interaction rule base and the drug use risk assessment model to check for drug incompatibilities. It assesses the medication risk and outputs intervention suggestions through a drug incompatibilities quantification model that integrates patient characteristics. Based on the patient's level of urgency and medication time window, it uses a dynamic priority model for emergency scenarios to calculate drug sorting and early warning priorities. The drug sorting and distribution execution module: after receiving intelligent decision instructions, controls the intelligent medicine cabinet and automatic sorting machine to complete drug sorting, and then distributes the drugs through the transmission track and AGV robot, and records the data of the entire sorting and distribution process; The medication monitoring and feedback module connects to the monitoring system and LIS system to collect vital signs and laboratory data of high-risk patients. Combined with subjective reaction data, it uses an anomaly recognition model to identify medication abnormalities and trigger graded early warnings. At the same time, it traces back the medication process to generate reports and supports medical staff annotations and adjustments. The system optimization and iteration module updates the drug information database daily, adds new contraindication cases and adjusts inventory, calls up full-process data weekly, optimizes model parameters using parameter optimization algorithms, and verifies and updates the system to the production environment after iteration.

2. The intelligent medication management and precise dispensing system for an emergency resuscitation room according to claim 1, characterized in that, The intelligent medication decision-making module contains the following content in its drug interaction rule base: Basic drug combination contraindications: chemical incompatibility, drug antagonistic combinations, combinations with superimposed toxic and adverse reactions, and combinations with conflicting metabolic pathways; Patient-specific drug interaction contraindications: These include contraindications related to a history of allergies and contraindications related to underlying diseases; Special contraindications for medication in emergency settings: contraindications related to time windows and contraindications for medication administered through multiple channels.

3. The intelligent medication management and precise dispensing system for emergency resuscitation rooms according to claim 1, characterized in that, In the intelligent medication decision-making module, the calculation formula for the drug use risk assessment model is as follows: ,in, It is a drug use risk assessment value, used to quantify the risk of drug use. It is an index of the patient's vital signs parameters, from 1 to... , representing different vital signs indicators, It is the total number of the patient's vital signs parameters. It is the weight of the patient's vital signs. These are normalized values ​​of vital signs. No. The raw values ​​of the patient's vital signs, The patient's medical history is a "risk signal": allergy history = 1, no allergy = 0. No. Patient medical history information, It is an index of patient medical history and risk factors, from 1 to... , indicating different types of medical history It is the total number of risk factors in the patient's medical history. It is the correlation between medical history and current medications. This is information about the drugs currently in use. It is the time decay coefficient. It is the time difference between the current time and the time of the most recent valid vital sign data collection, combined with the drug interaction rule base to set the drug use risk assessment threshold. ,when An alert will be triggered if the threshold is exceeded.

4. The intelligent drug management and precise distribution system for emergency resuscitation rooms according to claim 1, characterized in that, The calculation formula for constructing the drug incompatibility quantification model in the intelligent medication decision-making module is as follows: , This is a quantitative value for drug incompatibilities, used to quantify the degree of contraindication when two drugs are used together. This involves information on two drugs whose drug incompatibilities are to be evaluated. It is based on graph neural network (GNN) computation. and The interaction values ​​between two drugs reflect the risk of drug incompatibility. Graph Neural Networks (GNNs) employ a multi-relationship heterogeneous graph structure, where nodes represent drug chemical components / targets and edges represent interaction types. Embedding vectors are obtained through pre-training on a pharmaceutical database. It is an index of individual patient characteristic parameters, from 1 to... , representing different individual characteristics It is the total number of individual patient characteristic parameters. No. Patient individual characteristics information, No. Individual characteristics of patients and Factors influencing drug incompatibilities reflect the impact of individual differences on incompatibilities, and quantitative thresholds for drug incompatibilities are set based on clinical data. ,when Trigger an early warning and provide intervention suggestions.

5. The intelligent drug management and precise distribution system for emergency resuscitation rooms according to claim 4, characterized in that, The specific content of the intervention suggestions in the intelligent medication decision-making module is as follows: prioritize recommending equivalent alternative drugs with different mechanisms of action; when alternatives are not possible, adjust the drug dosage, interval or route of administration to reduce risk; clarify high-frequency monitoring indicators and emergency drugs and procedures; and provide personalized matching suggestions based on individual patient differences.

6. The intelligent medication management and precise dispensing system for an emergency resuscitation room according to claim 1, characterized in that, In the intelligent medication decision-making module, the calculation formula for constructing the dynamic priority model for emergency scenarios is as follows: ,in, This is a dynamic priority value for emergency scenarios, used to determine the priority of drug sorting and early warning. It contains patient information, including real-time vital signs data. It is based on real-time vital sign data from the patient information collection module. M represents drug information, including the drug's time window requirements. TimeWindow(M) is determined by the drug time window requirements in the drug information database, combined with the remaining time until the appropriate time to administer the medication. It is an index of sudden emergencies, from 1 to... This indicates different types of sudden emergencies. It is the total number of sudden emergencies. It is the first Emergency information When the system detects a sudden emergency, it assigns a value according to preset rules. Sort by numerical value The higher the value, the higher the priority of drug sorting and early warning.

7. The intelligent drug management and precise distribution system for emergency resuscitation rooms according to claim 1, characterized in that, In the medication monitoring and feedback module, the anomaly identification model is calculated using the following formula: ,in, These are medication anomaly detection values ​​used to determine whether a patient's condition becomes abnormal after medication administration. It refers to the time point after medication is administered. These are changes in physical signs after medication. This is the baseline of physical signs before medication. This is an expected change in the patient's condition. This is the baseline expectation of the condition. It is a risk assessment value for drug use.

8. The intelligent medication management and precise dispensing system for an emergency resuscitation room according to claim 7, characterized in that, The abnormality level classification in the medication monitoring and feedback module is as follows: when When this occurs, a Level 1 warning is triggered, reminding medical staff to pay attention to the patient and continuously monitor changes in vital signs, without requiring proactive adjustments to medication. when When this is triggered, a level-two alert is issued, reminding medical staff to quickly check the patient's condition, review the medication regimen, and assess whether to adjust the dosage or change the medication. when When an emergency alert is triggered, medical staff immediately rush to the patient's side, initiate emergency response procedures, and simultaneously trace back the entire medication process to investigate the root cause.

9. The intelligent drug management and precise distribution system for an emergency resuscitation room according to claim 7, characterized in that, In the system optimization iteration module, the model parameters are optimized using a parameter optimization algorithm, and the calculation formula is as follows: ,in, It is the optimized set of model parameters. It is the set of model parameters before optimization. It's the learning rate. These are the indexes of the sample data, from 1 to N, representing different sample data. It is the total number of sample data used for model parameter optimization. These are gradient values, representing the drug use risk assessment values ​​based on this sample. and actual results The calculations reflect the direction and magnitude of the model parameter adjustments. No. Drug use risk assessment value for each sample, No. The actual medication results of each sample It is a minimum value.

10. A method for intelligent drug management in an emergency resuscitation room, the method being used in the intelligent drug management and precise dispensing system for an emergency resuscitation room as described in any one of claims 1-9, characterized in that, The specific steps of this management method are as follows: S100, Information Collection and Preprocessing: This system connects to the hospital's HIS and EMR systems to collect basic patient information. It utilizes multi-parameter monitors and trauma assessment terminals to collect dynamic vital signs and medication time windows, and performs de-identification processing. A drug information database is built and loaded with drug instructions, clinical guidelines, and inventory data. Simultaneously, the collected data undergoes preprocessing and standardization. S200, Intelligent Medication Decision-Making: When doctors prescribe medication or nurses collect medication, it calls up a drug interaction rule base that includes contraindications to basic drug combinations, patient-specific contraindications, and special contraindications for emergency scenarios. It calculates the risk value by combining the drug use risk assessment model, evaluates the degree of contraindication by using a drug compatibility quantification model, generates intervention suggestions, and calculates drug sorting and early warning priorities based on the patient's critical condition, medication time window, and sudden emergency events through a dynamic priority model for emergency scenarios. S300, Drug Sorting and Distribution Execution: Receives intelligent decision-making instructions, controls intelligent medicine cabinets and automatic sorting machines to complete drug sorting, handles drugs with special storage conditions as required, distributes drugs through transmission tracks or AGV robots, and records the entire process data of sorting time, drug information, distribution path, equipment status and operators. S400, Medication Monitoring and Feedback: The medication monitoring and feedback module connects to the monitoring system and LIS system to collect vital signs, laboratory data and subjective reaction data of high-risk patients. It uses an anomaly recognition model to calculate anomaly recognition values, classifies the warnings into basic, intermediate and emergency levels according to the value and triggers corresponding measures. At the same time, it traces back the medication process to generate reports and supports medical staff annotations and adjustments. S500, system optimization and iteration: daily updates to the drug information database, addition of new contraindication cases, adjustment of inventory, and updates to instructions and guidelines; weekly access to full-process data, optimization of model parameters using parameter optimization algorithms, iteration verification, and updates to the production environment.

Images