Community hospital multi-source prescription intelligent examination and decoction preparation cooperation method and system

Abstract

The invention relates to the field of traditional Chinese medicine informatization, and provides a community hospital multi-source prescription intelligent prescription checking and decocting cooperation method and system, and the method comprises the steps: obtaining multi-source prescription data, including community hospital prescription pictures, HIS system prescription data and independent database prescription data; based on traditional Chinese medicine classical famous prescriptions, drug incompatibility and dosage standards, a knowledge base is constructed for intelligent prescriptions examination; determining a dispensing mode in a full-automatic mode, a semi-automatic mode and a manual dispensing mode according to the order quantity, the complexity and the equipment state, and designing a three-level backout mechanism; the processes of automatic barrel binding, medicine soaking and medicine decocting are executed; full-process monitoring is realized through an applet and a PC terminal; inventory management and automatic replenishment are realized based on an enterprise ERP system, the problems of single data source, lack of intelligent prescription checking, low dispensing efficiency, insufficient medicine decocting process monitoring, difficulty in inventory management and the like in an existing traditional Chinese medicine decoction and dispensing system are solved, and efficiency, accuracy, safety and the like of traditional Chinese medicine services are remarkably improved.

Description

Technical Field

[0001] The present invention relates to the field of information technology of traditional Chinese medicine, and specifically to a method and system for intelligent prescription review and decoction coordination of multi-source prescriptions in community hospitals, which is used to solve problems such as multi-data source processing, intelligent prescription review, automated adjustment and decoction management in traditional Chinese medicine decoction and preparation systems. Background Art

[0002] With the popularization and development of Traditional Chinese Medicine (TCM) services, demand for TCM decoction and dispensing services is growing. However, existing TCM decoction and dispensing systems face numerous challenges: First, existing systems typically only connect to a single hospital and process a single data source, failing to meet the interoperability needs of diverse medical institutions. Second, they lack a built-in mechanism for assisting with drug contraindication review, making it difficult to ensure the safety and rationality of prescriptions. Third, they lack real-time monitoring of the entire decoction and dispensing process, resulting in a poor user experience. Furthermore, most systems lack both semi-automated and automated dispensing solutions, making it difficult to flexibly adjust processing methods as order volume increases, leading to high labor costs. Finally, they lack effective integration with drug management systems, making drug price maintenance difficult and invoicing time-consuming and labor-intensive. Furthermore, the lack of automated replenishment and inventory management functions often leads to inventory errors.

[0003] These problems seriously restrict the efficiency and quality of TCM decoction and preparation services, and cannot meet the growing demand for TCM services. Therefore, there is an urgent need for a TCM decoction and preparation system that can solve the above problems. Summary of the Invention

[0004] The purpose of the present invention is to provide a method and system for intelligent prescription review and decoction coordination of multi-source prescriptions in community hospitals, aiming to solve the problems existing in the existing Chinese medicine decoction and preparation system, such as single data source, lack of intelligent prescription review, low adjustment efficiency, insufficient decoction process monitoring, and difficult inventory management.

[0005] The present invention proposes a method for intelligent prescription review and collaborative decoction of multi-source prescriptions in community hospitals, including:

[0006] Obtain multi-source prescription data, including prescription images from community hospitals, prescription data from medical institution HIS systems, and prescription data from medical institutions with independent databases;

[0007] Performing intelligent prescription review on the multi-source prescription data, including:

[0008] Based on the knowledge base of classic Chinese medicine prescriptions, drug incompatibility and dosage standards, the multi-source prescription data is reviewed for safety and rationality;

[0009] Based on the review results, determine the approval status of the prescription and send a reminder message to the relevant medical institution;

[0010] Perform prescription dispensing, including:

[0011] Based on the current order volume, order complexity and equipment status, determine the adjustment mode among the three adjustment modes: fully automatic adjustment, semi-automatic adjustment and manual adjustment;

[0012] Execute drug dispensing according to the determined dispensing mode, and if any abnormality occurs during the dispensing process, perform the withdrawal in the order of fully automatic dispensing → semi-automatic dispensing → manual dispensing;

[0013] Perform decoction treatment, including:

[0014] Automatically bind barrels, soak and decoct medicines for the prepared medicines;

[0015] Generate drug packaging and delivery information based on the decoction results;

[0016] Perform full-process monitoring and management, including:

[0017] Display the entire process of prescriptions from receipt to delivery through mini-programs and PC terminals;

[0018] Based on the enterprise ERP system, drug inventory management and automatic replenishment are realized.

[0019] Preferably, the step of obtaining multi-source prescription data specifically includes:

[0020] For prescription images from community hospitals, we use OCR recognition technology to extract key prescription information and generate standardized prescription data;

[0021] For prescription data from the HIS system of medical institutions, prescription information is received through HTTPS request POST interface and data standardization is performed;

[0022] For medical institutions with independent databases, prescription data is obtained through timed incremental capture and converted into a standard format;

[0023] The obtained multi-source prescription data is uniformly converted into a unified data structure including a basic information field group, a patient information field group, a diagnosis information field group, a prescription information field group, a drug information field group, a decoction information field group and a delivery information field group.

[0024] Preferably, the step of performing intelligent prescription review on multi-source prescription data specifically includes:

[0025] Collect traditional Chinese medicine literature data, modern pharmacology data, drug instructions data, clinical medication guidelines and drug adverse reaction data from multi-source knowledge data to build a drug knowledge base;

[0026] Conducting basic compliance checks on the multi-source prescription data, including prescription integrity checks, physician qualification checks, and prescription timeliness checks;

[0027] Conducting basic drug review on the multi-source prescription data, including drug compliance check, drug incompatibility check, and single drug dosage check;

[0028] Conduct patient relevance review on the multi-source prescription data, including age-related contraindication checks, gender-related contraindication checks, and special physical contraindication checks;

[0029] Conducting a disease relevance review on the multi-source prescription data, including checking the correlation between diagnosis and prescription, evaluating drug-certificate consistency, and checking for possible drug contraindications;

[0030] Based on the review results, prescriptions are divided into four levels: normal pass, warning pass, review required and rejection, and a corresponding processing mechanism is generated.

[0031] Preferably, the step of determining the adjustment mode among the three adjustment modes of fully automatic adjustment, semi-automatic adjustment and manual adjustment specifically includes:

[0032] When the current order volume is less than 30% of the equipment's processing capacity, the fully automatic adjustment mode is preferred;

[0033] When the current order volume is between 30% and 80% of the equipment's processing capacity, select the mode of using fully automatic and semi-automatic adjustment in parallel;

[0034] When the current order volume is greater than 80% of the equipment's processing capacity, the three modes of fully automatic adjustment, semi-automatic adjustment and manual adjustment are used in parallel;

[0035] For orders with special requirements or complex prescriptions, they are directly assigned to semi-automatic dispensing or manual dispensing mode;

[0036] By real-time monitoring of the working status of each adjustment line, the order allocation ratio can be dynamically adjusted.

[0037] Preferably, the execution steps of the fully automatic adjustment mode include:

[0038] Convert prescription information into instructions recognizable by automated dispensing equipment;

[0039] Check the location mapping and inventory status of required drugs in the automatic equipment;

[0040] Send instruction sequences to automatic dispensing equipment and monitor execution status in real time;

[0041] Verify the adjustment results through weight verification, image recognition verification and drug inventory verification;

[0042] When abnormal situations such as missing drugs, equipment failure or verification failure occur, the withdrawal mechanism is triggered.

[0043] Preferably, the execution steps of the semi-automatic adjustment mode include:

[0044] Break down prescriptions into dispensing task orders and assign them to operators;

[0045] Push task information to operators via PDA devices, including guidance information such as drug location and dosage;

[0046] The operator takes the medicine according to the PDA instructions and confirms the identity of the medicine by scanning the code;

[0047] Use smart electronic scales to detect the weight of medicines in real time to ensure accurate dosage;

[0048] After the operation is completed, take a photo and save it, and submit it to the system record;

[0049] When operational abnormalities, verification failures, or personnel interruptions occur, the withdrawal mechanism is triggered.

[0050] Preferably, the step of performing the decoction process specifically includes:

[0051] Assign a unique identifier to each medicine package and establish a binding relationship with the decoction container;

[0052] Set drug soaking parameters according to prescription characteristics, including water temperature, soaking time and water volume;

[0053] Select the decoction mode according to the prescription type, set the decoction time, temperature curve and special processing method;

[0054] The whole process of decoction is monitored by temperature, liquid level and video to ensure the quality of decoction;

[0055] For prescriptions that require multiple rounds of decoction, the system automatically controls the decoction process and storage of the medicinal solution;

[0056] After the decoction is completed, the liquid medicine is filtered, bagged and packaged, and quality inspection is carried out;

[0057] Through RFID / QR code recognition technology, decoction containers are intelligently scheduled in the conveyor system.

[0058] Preferably, the steps of performing full-process monitoring and management specifically include:

[0059] Set differentiated permissions for different user types, including patient users who can only view their own orders, medical institution users who can view all orders of the institution, and administrator users who can view all orders and system status;

[0060] The mini program displays the prescription review status, dispensing progress, decoction status, delivery information, and estimated completion time in real time.

[0061] Display multi-dimensional data such as order overview, prescription analysis, production monitoring and quality control in the PC management system;

[0062] Generate real-time reminders for abnormal situations in the process and provide guidance on handling the process;

[0063] Record key node data during system operation for efficiency analysis, quality problem analysis and cost analysis.

[0064] Preferably, the steps of implementing drug inventory management and automatic replenishment based on the enterprise ERP system specifically include:

[0065] Categorize and manage drug inventory and set up upper and lower limit warning mechanisms for inventory;

[0066] Record drug in-and-out information by scanning codes to achieve batch management and traceability;

[0067] Generate drug demand forecasts based on historical data analysis, trend analysis, and special factors;

[0068] Automatically generate replenishment plans based on forecast results and actual inventory status;

[0069] Establish data exchange with pharmaceutical companies' ERP systems through HTTPS request interfaces to achieve order status tracking and inventory data synchronization;

[0070] Detect abnormal situations in the data exchange process and provide automatic error correction or manual intervention mechanisms.

[0071] Intelligent prescription review and decoction coordination for multi-source prescriptions in community hospitals, including:

[0072] Prescription access module, used to obtain multi-source prescription data, including prescription images from community hospitals, prescription data from medical institution HIS systems, and prescription data from medical institutions with independent databases;

[0073] An intelligent prescription review module is used to review the safety and rationality of multi-source prescription data based on a knowledge base of classic Chinese medicine prescriptions, drug compatibility contraindications, and dosage standards. Based on the review results, it determines the approval status of the prescription and sends a prompt message to the relevant medical institution;

[0074] The dispensing management module is used to determine the dispensing mode among the three modes of fully automatic dispensing, semi-automatic dispensing, and manual dispensing based on the current order volume, order complexity, and equipment status. It executes drug dispensing according to the determined dispensing mode and, if an abnormality occurs during the dispensing process, performs the withdrawal in the order of fully automatic dispensing → semi-automatic dispensing → manual dispensing.

[0075] The decoction management module is used to automatically perform the barrel binding, automatic medicine soaking and decoction processes for the completed medicines, and generate medicine packaging and delivery information based on the decoction results;

[0076] The monitoring and viewing module is used to display the status of the entire process from prescription receipt to delivery through the mini program and PC;

[0077] ERP integration module for drug inventory management and automatic replenishment based on the enterprise ERP system.

[0078] The beneficial effects of the present invention include:

[0079] 1. Through the multi-source prescription access solution, unified access and processing of prescription data from different types of medical institutions such as community hospitals, large and medium-sized Chinese medicine hospitals, offline Chinese medicine clinics, private Chinese medicine clinics and online Internet Chinese medicine hospitals are achieved, breaking the data silos and improving the applicability and practicality of the system.

[0080] 2. Through the intelligent prescription review module, a knowledge base based on classic Chinese medicine prescriptions, drug compatibility contraindications and dosage standards has been established to conduct multi-dimensional safety and rationality reviews of prescriptions, significantly reducing the prescription error rate and potential risk of adverse drug reactions.

[0081] 3. Through multi-mode adjustment management, flexible switching between three modes of full-automatic adjustment, semi-automatic adjustment and manual adjustment is achieved, and a three-level withdrawal mechanism is designed, which greatly improves the adjustment efficiency and accuracy and reduces labor costs.

[0082] 4. Through automated decoction management, the entire process from tying barrels, soaking medicine to decocting medicine is automated, which improves the stability and consistency of decoction quality and reduces manual intervention.

[0083] 5. Through the full-process monitoring system, patients and medical institutions are provided with real-time status viewing capabilities for the entire process from prescription receipt to delivery, improving service transparency and user experience.

[0084] 6. By connecting with the enterprise ERP system, intelligent management and automatic replenishment of drug inventory are realized, solving the problem of inaccurate inventory counting and improving supply chain efficiency.

[0085] Overall, the present invention has built a complete intelligent solution for the decoction and preparation of traditional Chinese medicine, which has greatly improved the efficiency, accuracy and safety of traditional Chinese medicine services, reduced labor costs, improved user experience, and provided strong support for the modernization and standardization of traditional Chinese medicine services. BRIEF DESCRIPTION OF THE DRAWINGS

[0086] Figure 1 The process of the community hospital multi-source prescription intelligent prescription review and decoction collaborative method and system of the present invention Figure 1 .

[0087] Figure 2 The optimization process of the community hospital multi-source prescription intelligent prescription review and decoction collaborative method and system of the present invention Figure 2 .

[0088] Figure 3 The optimization process of the community hospital multi-source prescription intelligent prescription review and decoction collaborative method and system of the present invention Figure 3 . DETAILED DESCRIPTION

[0089] Please refer to Figure 1-Figure 3 The present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0090] Reference Figure 1 and Figure 2 The present invention provides a method for intelligent prescription review and collaborative decoction of multi-source prescriptions in community hospitals, comprising the following steps:

[0091] In one embodiment of the present invention, the system first needs to acquire prescription data from multiple sources. This data comes from a variety of sources, including prescription images from community hospitals, prescription data from medical institution HIS systems, and prescription data from medical institutions with independent databases. This multi-source data access capability is one of the core innovations of the present invention, overcoming the limitation of traditional systems that can only process single data sources.

[0092] For prescription images from community hospitals, this invention uses optical character recognition (OCR) technology to extract key prescription information. Specifically, the system first preprocesses the uploaded prescription image, including image denoising, tilt correction, and contrast enhancement, to improve subsequent recognition accuracy. After preprocessing, the system uses a deep learning model for text recognition, preferably an OCR model based on a convolutional neural network and an attention mechanism. This model is specifically trained for Traditional Chinese Medicine prescription formats and has a recognition accuracy of over 95%. After recognition, the system extracts key information and generates standardized prescription data.

[0093] For prescription data from a medical institution's HIS system, this system receives prescription information via an HTTPS POST request interface. To ensure data transmission security, the system uses the OAuth 2.0 authorization mechanism and the TLS 1.3 encryption protocol. After receiving the data, the system performs data standardization, including field mapping, encoding conversion, and unit unification.

[0094] For medical institutions with independent databases, this invention acquires prescription data through timed incremental capture. The system establishes data capture rules and implements incremental capture based on timestamps or sequence numbers. The preferred capture frequency is once every 5 minutes, which ensures the timeliness of data while avoiding excessive pressure on the source database.

[0095] Finally, the system converts the multi-source prescription data into a standard data structure, which includes the following field groups:

[0096] 1. Basic information field group: order ID, source hospital, creation time, status, etc.;

[0097] 2. Patient information field group: patient ID, name, age, gender, contact information, etc.;

[0098] 3. Diagnosis information field group: primary diagnosis, auxiliary diagnosis, symptom description, etc.;

[0099] 4. Prescription information field group: prescription type, prescribing physician, prescription time, prescription number, etc.;

[0100] 5. Drug information field group: drug list (drug name, specifications, dosage, usage, etc.);

[0101] 6. Decoction information field group: decoction method, dosage, special requirements, etc.

[0102] 7. Delivery information field group: delivery address, contact person, delivery time, etc.

[0103] This unified data structure ensures the standardization and normalization of subsequent processing procedures and is the basis for collaborative processing of multi-source heterogeneous data.

[0104] In another embodiment of the present invention, intelligent prescription review is a key step in ensuring the safety and rationality of prescriptions. This system first collects data from traditional Chinese medicine classics, modern pharmacology data, drug instructions, clinical medication guidelines, and adverse drug reaction data from multi-source knowledge data to build a comprehensive drug knowledge base.

[0105] The knowledge base is constructed using a structured approach, systematically organizing basic drug information, inter-drug relationships, contraindications, and dosage standards. Preferably, this system uses a graph database to store drug knowledge relationships, facilitating rapid query and reasoning. The knowledge base is updated regularly, typically monthly, and includes an emergency knowledge push mechanism to ensure the system can promptly respond to newly discovered drug safety issues.

[0106] Based on the constructed knowledge base, the system conducts multi-level and multi-dimensional review of prescriptions:

[0107] First, a basic compliance check is performed, including a prescription integrity check, a physician qualification check, and a prescription validity check. The default validity period for the prescription validity check is 3 days, which can be adjusted based on the medical institution's settings.

[0108] Next, a basic drug review is conducted, including drug compliance checks, drug compatibility checks, and single-drug dosage checks. During the drug compatibility check, the system focuses on traditional Chinese medicine taboos such as the 18 Antidotes and 19 Fears. Examples include the use of Chuanwu, Caowu, and Licorice together and the use of sulfur, fear, and potassium nitrate. During the single-drug dosage check, the system compares the prescribed drug dosage with the standard dosage and triggers an alert if it exceeds 120% of the usual dose.

[0109] Then, a patient-related review is conducted, including age-related, gender-related, and specific physical condition contraindications. For example, for pregnant women, the system will automatically flag prescriptions containing prohibited herbs such as "aconite" and "leech." For children, the system will check for inappropriate herbs such as "musk" and "bezoar."

[0110] Finally, a symptom relevance review is conducted, including checking the correlation between the diagnosis and prescription, assessing drug-certificate compatibility, and checking for possible drug contraindications. The system maps the diagnosis to the recommended medication and assesses the degree of match between the prescription and diagnosis. If the match falls below 70%, an alert is triggered.

[0111] Based on the above review results, the system divides prescriptions into four levels:

[0112] 1. Pass normally: No risk, go directly to the next process;

[0113] 2. Warning Pass: There are low-risk issues, a warning message is generated but the process is allowed to continue;

[0114] 3. Review Required: Moderate risk, requiring manual review and confirmation;

[0115] 4. Reject: There is a serious risk and the matter will be directly rejected.

[0116] For prescriptions with warning or review levels, the system automatically generates a notification and sends it to the designated responsible person at the relevant medical institution. The notification includes a description of the issue, risk level, and recommended actions. Medical institutions can access detailed review reports through the system's interface and make changes or force approvals. All review processes and results are logged in detail to support subsequent audit traceability.

[0117] After obtaining prescription data and completing intelligent prescription review, the present invention enters the dispensing phase. The innovation of the dispensing phase lies in the intelligent scheduling and seamless switching between three modes: fully automatic dispensing, semi-automatic dispensing, and manual dispensing.

[0118] The system dynamically determines which dispatch mode to use based on current order volume, order complexity, and equipment status. Specifically, when the current order volume is less than 30% of the equipment's processing capacity, the system prioritizes fully automatic dispatch, leveraging the efficiency of automated equipment. When the current order volume is between 30% and 80% of the equipment's processing capacity, the system utilizes both fully automatic and semi-automatic dispatch, balancing efficiency and flexibility. When the current order volume exceeds 80% of the equipment's processing capacity, the system utilizes all three dispatch modes in parallel to maximize processing capacity.

[0119] For orders with special requirements or complex prescriptions, the system directly allocates them to semi-automatic or manual dispensing modes to ensure accurate processing. Furthermore, the system dynamically adjusts order allocation ratios and optimizes resource utilization by monitoring the operating status of each dispensing route in real time.

[0120] In fully automatic dispensing mode, the system first converts prescription information into instructions recognizable by the automatic dispensing equipment. During this conversion process, the system checks the mapping of the required medications to the automatic dispensing equipment and the inventory status to ensure that all medications are properly accessible. The system then issues a sequence of instructions to the automatic dispensing equipment and monitors its execution status in real time. After the dispensing is completed, the system verifies the dispensing results through multiple methods, including weight verification, image recognition verification, and drug inventory verification, to ensure accuracy. In the event of anomalies such as missing medications, equipment failure, or verification failure, the system triggers a rollback mechanism, shifting the task to semi-automatic dispensing mode.

[0121] In semi-automatic dispensing mode, the system breaks down prescriptions into dispensing task orders and assigns them to operators. Task information, including guidance on medication location, dosage, and more, is pushed to operators via PDAs. Operators follow the PDA instructions to dispense medications and confirm the medication's identity by scanning a QR code. The system uses an intelligent electronic scale to measure the weight of dispensed medications in real time, maintaining an error range of ±5% to ensure accurate dosing. Upon completion, the system requires the operator to take a photo for record keeping and submit it to the system for record keeping. In the event of an operational anomaly, verification failure, or human interruption, the system triggers a rollback mechanism, transferring the task to manual dispensing mode.

[0122] In manual dispensing mode, the system generates a standardized dispensing form containing complete drug information, dosage, and special requirements. Dispensers pick up medications according to the form, with the entire process recorded by surveillance cameras and photographed at key points. Upon completion, a second person independently verifies the dispensing process, confirming both weight and appearance to ensure accuracy.

[0123] The present invention's downgrade mechanism is a key design element in ensuring reliable dispatching. When a problem arises in a high-level dispatch mode, the system automatically downgrades the dispatching process: from fully automatic to semi-automatic, and then to manual dispatching. The reason for each downgrade is recorded to facilitate subsequent analysis and system optimization.

[0124] After the adjustment is completed, the invention enters the decoction process. The innovation of the decoction process is to realize the full process automation control from tying the barrel, soaking the medicine to decoction.

[0125] First, the system assigns a unique identifier to each medicine package and establishes a binding relationship with the decoction container. The identifier is a QR code containing information such as the order number and the medicine package number. The binding process is confirmed by scanning both the medicine package identifier and the decoction container identifier simultaneously to ensure a one-to-one correspondence. Furthermore, the system uses a camera for visual verification to ensure the binding is accurate.

[0126] Next, the system sets the drug soaking parameters based on the prescription characteristics. The water temperature setting range is 60-100°C and can be adjusted according to the properties of the medicinal materials. For example, cooling medicinal materials generally use a water temperature of 60-70°C, and hot medicinal materials generally use a water temperature of 80-90°C. The soaking time setting range is 30 minutes to 2 hours, determined by the hardness and properties of the medicinal materials. For example, hard medicinal materials (such as shells) have a longer soaking time, generally 1.5-2 hours, while floral and leafy medicinal materials have a shorter soaking time, generally 30-45 minutes. The amount of water is automatically calculated based on the proportion of the medicinal materials, generally following the principle of a 1:10 ratio of medicinal materials to water.

[0127] The system then selects a decoction mode based on the prescription type, setting the decoction time, temperature curve, and any special handling methods. Common decoction modes include standard decoction and slow decoction. In standard decoction mode, the initial decoction temperature is 100°C for 20 minutes, then lowered to 95°C for 40 minutes. In slow decoction mode, the temperature is maintained at 85-90°C throughout, extending the decoction duration to 1.5-2 hours. For special requirements, such as "decoction first" or "add later," the system adjusts the decoction process accordingly. For example, herbs marked "decoction first" are decocted 30 minutes in advance, while herbs marked "add later" are added in the last 15 minutes of the decoction process.

[0128] During the decoction process, the system monitors temperature, liquid level, and video in real time. Temperature monitoring uses a digital temperature sensor with an accuracy of ±0.5°C; liquid level monitoring uses an ultrasonic level sensor to prevent dry pot; and video monitoring uses a high-definition camera to record the entire decoction process, facilitating quality control and problem tracing.

[0129] For prescriptions requiring multiple decoctions, the system automatically manages the decoction process and storage of the medicinal solution. Typically, a Chinese medicine prescription requires two to three decoctions. The system automatically records the time and parameters for each decoction, stores the solution in a temporary container, and recombines the solution after all decoctions are complete.

[0130] After the decoction is complete, the system filters the liquid, bags it, and performs quality inspections. Filtration utilizes multiple layers of screens, filtering from coarse to fine, ensuring the liquid is free of impurities. Vacuum packaging technology is used for bagging to extend the shelf life of the liquid. Quality inspections randomly select 5% of the finished products for appearance, odor, and sealability checks to ensure product quality.

[0131] Finally, the system uses RFID / QR code recognition technology to intelligently schedule decoction containers within the conveyor system. The system calculates the optimal route based on order priority, processing stage, and target workstation, reducing wait times and improving production efficiency. The entire conveyor journey is recorded by video surveillance, with real-time node status reporting. If an anomaly such as a jam is detected, the system automatically switches to an alternate route or triggers manual intervention.

[0132] In another embodiment of the present invention, full-process monitoring and management is an important part of improving system transparency and user experience. This system displays the full process status of prescriptions from receipt to delivery through mini-programs and PC terminals, providing differentiated monitoring experiences for different users.

[0133] The system sets differentiated permissions for different user types: Patient users can only view their own orders, medical institution users can view all orders for their institution, and administrator users can view all orders and system status. This hierarchical permission system ensures data security while meeting the monitoring needs of different users.

[0134] On the mini-program, the system displays real-time prescription review status, dispensing progress, decoction status, delivery information, and estimated completion time. Users can click to view photos of key milestones and operation records, providing an intuitive understanding of order processing status. The system also provides push notifications for status changes, a feedback channel, and a customer service interface to enhance the user experience.

[0135] The PC-based management system displays multi-dimensional data, including order overview, prescription analysis, production monitoring, and quality control. Order overview includes quantity statistics, status distribution, and processing time analysis; prescription analysis includes drug usage frequency, exception statistics, and prescription source distribution; production monitoring includes equipment status, efficiency analysis, and load distribution; and quality control includes indicators such as exception rate, rejection rate, and customer satisfaction.

[0136] The system generates real-time alerts for process anomalies and provides guidance on how to handle them. For example, if a device failure rate exceeds 5%, the system sends an alert to the administrator and recommends maintenance. If a process takes an unusually long time (exceeding 150% of the average), the system automatically flags it and prompts intervention.

[0137] The system also records key operational data for efficiency, quality, and cost analysis. This data is presented in daily, weekly, and monthly reports, helping managers understand system performance, identify potential issues, and optimize operational strategies.

[0138] In the last core link of the present invention, the system realizes intelligent management and automatic replenishment of drug inventory by connecting with the enterprise ERP system, solving the problem of inaccurate inventory counting in traditional systems.

[0139] The system manages drug inventory by categorization, classifying it by drug type, frequency of use, and storage conditions for easier management and query. Furthermore, the system implements an inventory alert mechanism with upper and lower limits. When a drug's inventory falls below the lower limit (usually 30% of safety stock), a replenishment reminder is triggered. When it exceeds the upper limit (usually 90% of maximum inventory), a backlog alert is triggered to prevent capital tie-up.

[0140] Drug inbound and outbound shipments are recorded by scanning a QR code, enabling batch management and traceability. For each incoming batch, the system records information such as supplier, production date, and batch number, establishing a complete traceability chain to facilitate quality management and problem tracing. The system optimally supports the FIFO (first-in, first-out) principle, prioritizing older inventory and minimizing the risk of expiration.

[0141] The system generates drug demand forecasts based on historical data analysis, trend analysis, and special factor considerations. Historical data analysis primarily examines drug usage over the past three to six months to identify seasonal and cyclical variations; trend analysis uses time series models to predict future demand trends; and special factor considerations incorporate unconventional factors such as holidays and epidemics into the forecast model to improve forecast accuracy.

[0142] Based on forecast results and actual inventory levels, the system automatically generates a replenishment plan. This plan includes information such as drug name, specification, quantity, and recommended suppliers, and is prioritized based on urgency. For commonly used medicinal materials, the system uses an economic order quantity model to calculate the optimal order quantity, balancing ordering and inventory costs. In the event of an emergency out-of-stock situation, the system activates an emergency replenishment mechanism, prioritizing processing and ensuring uninterrupted production.

[0143] The system establishes data exchange with pharmaceutical companies' ERP systems via an HTTPS request interface, enabling order status tracking and inventory data synchronization. Data exchange utilizes a unified XML / JSON format, with clearly defined field mapping and data validation rules to ensure data consistency and accuracy. The interface design includes authentication and authorization mechanisms, encrypted transmission, and exception handling to ensure the security and reliability of data exchange.

[0144] During data exchange, the system detects anomalies and provides automatic correction or manual intervention mechanisms. When data inconsistencies are detected, the system automatically compares historical records to identify the cause and correct them. If automatic correction fails to resolve the issue, the system generates a manual intervention notice for dedicated personnel to handle. Detailed logs are kept for all data exchange processes to support subsequent audits and problem tracing.

[0145] Reference Figure 1 and Figure 2 The present invention also provides a multi-source prescription intelligent review and decoction collaborative system for community hospitals, including a prescription access module 1, an intelligent prescription review module 2, a dispensing management module 3, a decoction management module 4, a monitoring and viewing module 5 and an ERP integration module 6.

[0146] Prescription Access Module 1 is used to acquire multi-source prescription data, including prescription images from community hospitals, prescription data from medical institution HIS systems, and prescription data from medical institutions with independent databases. This module serves as the system's data entry point, processing prescription data from various sources and converting it into a unified standard format. Prescription Access Module 1 comprises three submodules: an OCR recognition engine, an API interface service, and a data capture service, each of which processes prescription data from different sources.

[0147] Intelligent Prescription Review Module 2 is used to review the security and rationality of multi-source prescription data based on a knowledge base constructed from classic Traditional Chinese Medicine formulas, drug compatibility contraindications, and dosage standards. Based on the review results, it determines the prescription's approval status and sends a notification message to the relevant medical institution. This module serves as the system's security safeguard, responsible for verifying the compliance and safety of prescriptions. Intelligent Prescription Review Module 2 includes three submodules: knowledge base management, multi-dimensional review, and result processing, completing the complete process from knowledge acquisition to review result processing.

[0148] Dispensing Management Module 3 determines the dispensing mode from among fully automatic, semi-automatic, and manual based on the current order volume, order complexity, and equipment status. It executes drug dispensing according to the determined dispensing mode and, if an exception occurs during the dispensing process, de-adjusts the order from fully automatic to semi-automatic to manual. This module serves as the system's execution center, responsible for accurately dispensing drugs according to prescription requirements. Dispensing Management Module 3 includes three submodules: mode scheduling, execution control, and exception handling, ensuring the efficiency and reliability of the dispensing process.

[0149] The decoction management module 4 automatically executes the barrel binding, infusion, and decoction processes for prepared medicines, and generates drug packaging and delivery information based on the decoction results. This module serves as the system's production center, responsible for decocting prepared herbs into liquid medicine. The decoction management module 4 comprises four submodules: barrel binding management, infusion control, decoction control, and delivery scheduling, automating and standardizing the decoction process.

[0150] Monitoring and Viewing Module 5 displays the entire prescription process, from receipt to delivery, through the mini-program and PC. This module serves as the system's display window, providing real-time visibility into order processing status for different users. Monitoring and Viewing Module 5 includes four submodules: User Rights Management, Status Display, Exception Alerts, and Data Analysis, meeting the monitoring needs of different users.

[0151] ERP Integration Module 6 implements pharmaceutical inventory management and automated replenishment within the enterprise ERP system. This module provides logistical support for the system, ensuring the continuity of pharmaceutical supply and accurate inventory management. ERP Integration Module 6 includes four submodules: inventory management, demand forecasting, replenishment planning, and data exchange, enabling seamless integration with the enterprise ERP system.

[0152] Standardized data interfaces enable seamless integration and information sharing between modules, forming a complete closed-loop system. Prescription data flows from Prescription Access Module 1 to Intelligent Prescription Review Module 2. After approval, it flows to Dispensing Management Module 3. After dispensing is complete, it flows to Decoction Management Module 4, ultimately creating the finished liquid medicine and entering the distribution process. Status information for the entire process is transmitted in real time to Monitoring and Viewing Module 5 for user review. Simultaneously, information on drug consumption during the dispensing and decoction processes is transmitted to ERP Integration Module 6 for inventory management and replenishment planning.

[0153] Through the collaborative work of six functional modules, the present invention constructs a complete intelligent solution for the decoction and preparation of traditional Chinese medicine, effectively solving the problems existing in the existing system, such as single data source, lack of intelligent prescription review, low dispensing efficiency, insufficient monitoring of the decoction process, and difficulty in inventory management, thus providing strong support for the modernization and standardization of traditional Chinese medicine services.

[0154] The above-described embodiments merely represent specific implementations of the present invention. While the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the present invention. It should be noted that a person skilled in the art would be able to make numerous variations and improvements without departing from the spirit of the present invention, and all such variations and improvements fall within the scope of protection of the present invention.

Claims

1. A method and system for intelligent prescription review and decoction coordination of multi-source prescriptions in community hospitals, characterized by: include: Obtain multi-source prescription data, including prescription images from community hospitals, prescription data from medical institution HIS systems, and prescription data from medical institutions with independent databases; Performing intelligent prescription review on the multi-source prescription data, including: Based on the knowledge base of classic Chinese medicine prescriptions, drug incompatibility and dosage standards, the multi-source prescription data is reviewed for safety and rationality; Based on the review results, determine the approval status of the prescription and send a reminder message to the relevant medical institution; Perform prescription dispensing, including: Based on the current order volume, order complexity and equipment status, determine the adjustment mode among the three adjustment modes: fully automatic adjustment, semi-automatic adjustment and manual adjustment; Execute drug dispensing according to the determined dispensing mode, and if any abnormality occurs during the dispensing process, perform the withdrawal in the order of fully automatic dispensing → semi-automatic dispensing → manual dispensing; Perform decoction treatment, including: Automatically bind barrels, soak and decoct medicines for the prepared medicines; Generate drug packaging and delivery information based on the decoction results; Perform full-process monitoring and management, including: Display the entire process of prescriptions from receipt to delivery through mini-programs and PC terminals; Based on the enterprise ERP system, drug inventory management and automatic replenishment are realized.

2. The method according to claim 1, characterized in that The step of obtaining multi-source prescription data specifically includes: For prescription images from community hospitals, we use OCR recognition technology to extract key prescription information and generate standardized prescription data; For prescription data from the HIS system of medical institutions, prescription information is received through HTTPS request POST interface and data standardization is performed; For medical institutions with independent databases, prescription data is obtained through timed incremental capture and converted into a standard format; The obtained multi-source prescription data is uniformly converted into a unified data structure including a basic information field group, a patient information field group, a diagnosis information field group, a prescription information field group, a drug information field group, a decoction information field group and a delivery information field group.

3. The method according to claim 1, characterized in that The steps of performing intelligent prescription review on multi-source prescription data specifically include: Collect traditional Chinese medicine literature data, modern pharmacology data, drug instructions data, clinical medication guidelines and drug adverse reaction data from multi-source knowledge data to build a drug knowledge base; Conducting basic compliance checks on the multi-source prescription data, including prescription integrity checks, physician qualification checks, and prescription timeliness checks; Conducting basic drug review on the multi-source prescription data, including drug compliance check, drug incompatibility check, and single drug dosage check; Conduct patient relevance review on the multi-source prescription data, including age-related contraindication checks, gender-related contraindication checks, and special physical contraindication checks; Conducting a disease relevance review on the multi-source prescription data, including checking the correlation between diagnosis and prescription, evaluating drug-certificate consistency, and checking for possible drug contraindications; Based on the review results, prescriptions are divided into four levels: normal pass, warning pass, review required and rejection, and a corresponding processing mechanism is generated.

4. The method according to claim 1, wherein The step of determining the adjustment mode among the three adjustment modes of fully automatic adjustment, semi-automatic adjustment and manual adjustment specifically includes: When the current order volume is less than 30% of the equipment's processing capacity, the fully automatic adjustment mode is preferred; When the current order volume is between 30% and 80% of the equipment's processing capacity, select the mode of using fully automatic and semi-automatic adjustment in parallel; When the current order volume is greater than 80% of the equipment's processing capacity, the three modes of fully automatic adjustment, semi-automatic adjustment and manual adjustment are used in parallel; For orders with special requirements or complex prescriptions, they are directly assigned to semi-automatic dispensing or manual dispensing mode; By real-time monitoring of the working status of each adjustment line, the order allocation ratio can be dynamically adjusted.

5. The method according to claim 1, wherein The execution steps of the fully automatic adjustment mode include: Convert prescription information into instructions recognizable by automated dispensing equipment; Check the location mapping and inventory status of required drugs in the automatic equipment; Send instruction sequences to automatic dispensing equipment and monitor execution status in real time; Verify the adjustment results through weight verification, image recognition verification and drug inventory verification; When abnormal situations such as missing drugs, equipment failure or verification failure occur, the withdrawal mechanism is triggered.

6. The method according to claim 1, characterized in that The execution steps of the semi-automatic adjustment mode include: Break down prescriptions into dispensing task orders and assign them to operators; Push task information to operators via PDA devices, including guidance information such as drug location and dosage; The operator takes the medicine according to the PDA instructions and confirms the identity of the medicine by scanning the code; Use smart electronic scales to detect the weight of medicines in real time to ensure accurate dosage; After the operation is completed, take a photo and save it, and submit it to the system record; When operational abnormalities, verification failures, or personnel interruptions occur, the withdrawal mechanism is triggered.

7. The method according to claim 1, characterized in that The steps of performing the decoction process specifically include: Assign a unique identifier to each medicine package and establish a binding relationship with the decoction container; Set drug soaking parameters according to prescription characteristics, including water temperature, soaking time and water volume; Select the decoction mode according to the prescription type, set the decoction time, temperature curve and special processing method; The whole process of decoction is monitored by temperature, liquid level and video to ensure the quality of decoction; For prescriptions that require multiple rounds of decoction, the system automatically controls the decoction process and storage of the medicinal solution; After the decoction is completed, the liquid medicine is filtered, bagged and packaged, and quality inspection is carried out; Through RFID / QR code recognition technology, decoction containers are intelligently scheduled in the conveyor system.

8. The method according to claim 1, characterized in that The steps of performing full-process monitoring and management specifically include: Set differentiated permissions for different user types, including patient users who can only view their own orders, medical institution users who can view all orders of the institution, and administrator users who can view all orders and system status; The mini program displays the prescription review status, dispensing progress, decoction status, delivery information, and estimated completion time in real time. Display multi-dimensional data such as order overview, prescription analysis, production monitoring and quality control in the PC management system; Generate real-time reminders for abnormal situations in the process and provide guidance on handling the process; Record key node data during system operation for efficiency analysis, quality problem analysis and cost analysis.

9. The method according to claim 1, characterized in that The steps of implementing drug inventory management and automatic replenishment based on the enterprise ERP system specifically include: Categorize and manage drug inventory and set up upper and lower limit warning mechanisms for inventory; Record drug in-and-out information by scanning codes to achieve batch management and traceability; Generate drug demand forecasts based on historical data analysis, trend analysis, and special factors; Automatically generate replenishment plans based on forecast results and actual inventory status; Establish data exchange with pharmaceutical companies' ERP systems through HTTPS request interfaces to achieve order status tracking and inventory data synchronization; Detect abnormal situations in the data exchange process and provide automatic error correction or manual intervention mechanisms.

10. The intelligent prescription review and decoction collaborative system for multi-source prescriptions in community hospitals is characterized by: include: Prescription access module, used to obtain multi-source prescription data, including prescription images from community hospitals, prescription data from medical institution HIS systems, and prescription data from medical institutions with independent databases; An intelligent prescription review module is used to review the safety and rationality of multi-source prescription data based on a knowledge base of classic Chinese medicine prescriptions, drug compatibility contraindications, and dosage standards. Based on the review results, it determines the approval status of the prescription and sends a prompt message to the relevant medical institution; The dispensing management module is used to determine the dispensing mode among the three modes of fully automatic dispensing, semi-automatic dispensing, and manual dispensing based on the current order volume, order complexity, and equipment status. It executes drug dispensing according to the determined dispensing mode and, if an abnormality occurs during the dispensing process, performs the withdrawal in the order of fully automatic dispensing → semi-automatic dispensing → manual dispensing. The decoction management module is used to automatically perform the barrel binding, automatic medicine soaking and decoction processes for the completed medicines, and generate medicine packaging and delivery information based on the decoction results; The monitoring and viewing module is used to display the status of the entire process from prescription receipt to delivery through the mini program and PC; ERP integration module for drug inventory management and automatic replenishment based on the enterprise ERP system.

Images