Medical insurance external prescription circulation and data supervision method
By constructing a pharmacy recommendation and risk identification model and combining it with blockchain technology, the secure transfer of hospital prescriptions and pharmacy information matching during online medical consultations have been achieved, solving the problems of prescription leakage and information security, and improving the efficiency of drug purchases and the security of medical insurance funds.
Patent Information
- Application Number
- CN202510972739.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-07-15
- Publication Date
- 2025-10-17
AI Technical Summary
Hospital prescriptions are easily leaked during Internet medical consultations, and the security of patients' personal information cannot be guaranteed. In addition, it is difficult to match information between hospitals and pharmacies, which leads to the obstruction of prescription outflow.
A pharmacy recommendation model and a risk identification model are constructed, and the hash value is stored on the blockchain. Combined with intelligent pharmacy recommendation and hierarchical interception processing, the immutability of the blockchain is used for prescription verification and circulation supervision, and information access is controlled by temporarily sharing passwords to generate drug delivery vouchers for quality evidence storage.
It enables precise pharmacy recommendations, ensures prescription information security, prevents tampering, protects patient privacy, improves medication purchase efficiency, ensures compliance with medical insurance settlement and drug quality, and enhances risk identification capabilities.
Smart Images
Figure CN120809056A_ABST
Abstract
Description
TECHNICAL FIELD
[0001] The present application relates to the technical field of smart medical treatment, and in particular to a medical insurance out-of-hospital prescription circulation and data supervision method. BACKGROUND
[0002] Under the premise that the state encourages the separation of medicine and pharmacy and reduces the proportion of hospital drugs, how do hospital prescriptions flow out and how do they better empower pharmacies to serve patients is undoubtedly an industry-level demand. Currently, the HIS system of the hospital and the ERP system of the pharmacy are relatively independent in the industry, and it is difficult for the hospital, pharmacy and other drug information to match, the outflow of the hospital's prescription is blocked, the pharmacy needs a prescription for the sale of prescription drugs, and the sale is blocked, with many obstacles, which are problems that need to be solved urgently. The medical insurance electronic prescription circulation refers to the process that after the insured person is treated in a designated medical institution, the doctor issues an electronic prescription and uploads it to the medical insurance platform, and then circulates it to the designated pharmacy, and the patient directly settles the medicine in the pharmacy with the medical insurance electronic certificate. The electronic prescription issued by the doctor usually includes the patient's diagnosis results and the patient's drug information, which involves the patient's personal privacy information. In the process of internet treatment, the electronic prescription is circulated and transmitted on the internet hospital platform and the internet pharmacy platform, which makes the electronic prescription prone to information leakage, which cannot guarantee the personal information security of the patient in the process of internet treatment. SUMMARY
[0003] The purpose of the present application is to provide a medical insurance out-of-hospital prescription circulation and data supervision method, which can solve the problem of electronic prescription information leakage and cannot guarantee the personal information security of the patient in the process of internet treatment.
[0004] Technical scheme: In order to solve the above technical problems, according to one aspect of the present application, more specifically, a medical insurance out-of-hospital prescription circulation and data supervision method, specifically comprising the following steps:
[0005] S1, constructing a pharmacy recommendation model and a risk identification model: obtaining multi-source historical data, inputting the processed historical data into the pharmacy recommendation model and the risk identification model to train the model;
[0006] S2, prescription generation and evidence anchoring: the doctor issues a patient's prescription through the HIS system, calculates the prescription hash value, and writes the prescription hash value into the blockchain;
[0007] S3, intelligent pharmacy recommendation and prescription risk interception: obtaining patient, pharmacy and supervision three-party historical data, the model performs intelligent pharmacy recommendation based on the current prescription information through the three-party historical data, and performs hierarchical interception processing on the current prescription through the model based on the prescription data and patient data;
[0008] S4, security authorization and password sharing: the patient performs security identification through the medical insurance platform, selects a pharmacy from the list of recommended pharmacies to share the prescription, and generates a temporary sharing password;
[0009] S5, password identification and payment settlement: the pharmacy identifies the patient's temporary sharing password to obtain the patient's prescription, verifies the patient's prescription related information, and performs medical insurance settlement after verification, and stores the settlement through the blockchain;
[0010] S6, drug delivery and quality storage: obtain the related data information of the drug to be delivered to generate a drug delivery voucher, and upload the delivery voucher to the blockchain;
[0011] S7, model retraining: after processing the related data of the external prescription, the model is input into the model for retraining and optimization.
[0012] Further, in step S1, the pharmacy recommendation model is constructed, which specifically includes the following steps:
[0013] S11, obtaining patient historical data and pharmacy historical data;
[0014] S12, extracting key features of patient historical data and key features of pharmacy historical data, obtaining patient key feature historical data and pharmacy key feature historical data;
[0015] S13, performing outlier processing, missing value filling on patient key feature historical data and pharmacy key feature historical data, and performing feature enhancement on data to obtain second feature historical data;
[0016] S14, performing feature classification on the second feature historical data and extracting time features to construct a training data set, dividing the training data set into a training set, a test set and a validation set according to a proportion, and training the pharmacy recommendation model.
[0017] Further, in step S1, when training the risk identification model, the risk level is divided into high level, medium level and low level, and high risk condition, medium risk condition and low risk condition are set respectively; any one of the high risk conditions is determined as high level risk, two or more medium risk conditions are determined as medium level risk, and only one low risk condition or no risk is determined as low level risk.
[0018] Further, in step S2, after the doctor issues the patient's prescription, the structured prescription data is standardized, the prescription hash value is calculated, and the prescription hash value is written into the blockchain as an initial storage anchor point.
[0019] Further, in step S3, the real-time features of the current prescription are input into the intelligent pharmacy recommendation model. The model generates a pharmacy recommendation score based on the historical data of patients, pharmacies, and regulators, as well as the real-time features of the current prescription. The pharmacies are ranked from high to low based on the recommendation score for the patient to choose.
[0020] Further, in step S3, based on the prescription data and patient data, the model identifies and classifies high-risk prescriptions. High-risk prescriptions are immediately frozen, and a freeze report is sent to multiple parties. The medical insurance verification is notified to intervene, and medium-risk prescriptions are marked as pending for manual review. Based on the review results, the audit conclusion is made to continue the circulation, upgrade to high-risk, or supplement materials. Prescriptions that are not reviewed within a certain time are automatically upgraded to high-risk for early warning. Low-risk prescriptions are kept circulating in real-time and are monitored in real-time. The complete circulation path of abnormal state prescriptions is recorded. Risk prescription feature data is extracted to determine whether it is a new pattern. If it is determined to be a new pattern, a new identification rule is created and transmitted to the risk identification model database and pushed to all nodes. Otherwise, the existing identification rules are strengthened.
[0021] Further, in step S5, when the pharmacy verifies the prescription, the hash value of the prescription obtained by the pharmacy is compared with the hash value of the prescription stored in the blockchain. If the two hash values are consistent, the verification is passed. Otherwise, the prescription is tampered with, triggering an alarm.
[0022] Further, in step S5, when the medical insurance settlement is completed, payment voucher data is constructed, and the hash value of the calculation voucher data is stored in the blockchain for medical insurance fund audit and pharmacy settlement to verify whether the hash value of the voucher data is consistent. If it is consistent, it means that the medical insurance settlement is compliant. Otherwise, it means that the medical insurance settlement is non-compliant, triggering a financial audit alarm.
[0023] Further, in step S6, the drug regulatory code and temperature sensor data of the cold chain drug are obtained when the drug is delivered. The drug delivery voucher is constructed based on the above data. The drug itself hash value and the environment hash value are calculated. The drug delivery hash value is obtained by combining the drug itself hash value and the environment hash value. The patient scans the delivery verification hash value after receiving the drug and compares it with the drug delivery hash value. If they are consistent, it means that the drug is not suspected to be replaced or stored irregularly. Otherwise, it means that the drug is suspected to be replaced or stored irregularly.
[0024] Beneficial effects:
[0025] 1. By acquiring multi-source historical data and training models, a reliable model foundation is provided for subsequent intelligent pharmacy recommendations and prescription risk identification. Processing of patient and pharmacy historical data, such as outlier handling, missing value filling, and feature enhancement, improves data quality, enabling the trained pharmacy recommendation model to more accurately match patients and pharmacies. The risk identification model, by categorizing risk levels and setting corresponding conditions, can more scientifically identify prescription risks, providing a basis for subsequent risk interception.
[0026] 2. After the doctor issues a prescription, he calculates the hash value and writes it into the blockchain. Taking advantage of the blockchain's tamper-proof nature, the prescription hash value is used as the initial evidence anchor point to ensure the originality and integrity of the prescription information, prevent the prescription from being maliciously tampered with, and provide a trusted basic certificate for subsequent prescription verification and circulation.
[0027] 3. Intelligent pharmacy recommendations based on third-party historical data and current prescription information can accurately recommend appropriate pharmacies to patients, saving them time and effort in choosing a pharmacy and improving drug purchasing efficiency. Prescriptions are intercepted and processed at a tiered level: high-risk prescriptions are immediately frozen and notified for review, medium-risk prescriptions are manually reviewed, and low-risk prescriptions are monitored in real time. This allows for timely identification and processing of prescriptions at different risk levels, effectively reducing medication risks and ensuring patient medication safety. Furthermore, the flow path of abnormal prescriptions is recorded and identification rules are updated to continuously improve risk identification capabilities.
[0028] 4. Patients are securely identified and generate temporary sharing passwords through the medical insurance platform. Only authorized pharmacies can obtain prescriptions through the passwords, reducing the risk of prescription information leakage during the sharing process, protecting patients' personal privacy and prescription information security, and ensuring the security and controllability of prescription sharing.
[0029] 5. Pharmacies use passwords to obtain prescriptions and verify the hash value, ensuring the authenticity and integrity of the prescription, ensuring the accuracy of prescription circulation. After medical insurance settlement, the payment voucher data hash value is stored on the blockchain for verification, ensuring the compliance of medical insurance settlements and preventing settlement data from being tampered with. This facilitates medical insurance fund audits and pharmacy settlement verification, ensuring the security of medical insurance funds.
[0030] 6. Obtain drug-related data to generate a delivery receipt and upload it to the blockchain. Patients can verify that the drug has not been swapped or stored illegally by comparing the hash value, ensuring drug quality and ensuring patient medication safety. Blockchain evidence storage enables traceability of drug delivery information, providing a reliable basis for tracing and determining responsibility for drug quality issues. By using the prescription content hash, payment receipt hash, and drug traceability hash, a three-chain mutual verification mechanism is implemented to build a complete evidence chain: prescription chain, payment chain, and drug chain.
[0031] 7. Use the relevant data of the current external dispensing prescription for model retraining, continuously optimize the pharmacy recommendation and risk identification model, make the model better adapt to the changes of the actual business scene, improve the recommendation accuracy and risk identification ability of the model, and continuously improve the efficiency and reliability of the entire medical insurance external dispensing prescription circulation and data supervision process. BRIEF DESCRIPTION OF DRAWINGS
[0032] Figure 1 is a method flow diagram. DETAILED DESCRIPTION
[0033] In order to make the technical scheme of the present application clearer, the present application will be further described in detail below in combination with the drawings and specific embodiments.
[0034] Embodiment 1
[0035] A medical insurance external dispensing prescription circulation and data supervision method, specifically comprising the following steps:
[0036] Step 1, build a pharmacy recommendation model and a risk identification model: obtain multi-source historical data, input the processed historical data into the pharmacy recommendation model and the risk identification model to train the model. The pharmacy recommendation model is built and specifically comprises the following steps:
[0037] 1. Obtain patient historical data and pharmacy historical data;
[0038] 2. Extract the key features of the patient historical data and the key features of the pharmacy historical data, including patient historical behaviors such as price sensitivity, delivery effectiveness preference, etc.; pharmacy services such as average compliance effectiveness, prescription error rate, etc.; supervision records such as the number of false settlements, unqualified drug inspection rate, etc.; drug circulation such as cold chain compliance rate, near expiration drug proportion, etc., to obtain patient key feature historical data and pharmacy key feature historical data;
[0039] 3. Perform outlier processing on the patient key feature historical data and the pharmacy key feature historical data, correct the outliers to 1.5 times the regional average value, fill in the missing values by using the time series interpolation method, and perform feature enhancement on the data to obtain second feature historical data;
[0040] 4. Classify the second feature historical data and extract time features to construct a training data set, divide the training data set into a training set, a test set and a validation set according to a ratio of 7:2:1, and train the pharmacy recommendation model.
[0041] In the training of the risk identification model, the risk level is divided into high, medium and low levels, and high-risk conditions are set, such as 7 days within the same patient with the same drug prescription ≥ 4 times, settlement price exceeds the medical insurance limit price by 30%, matching known fraud gang mode, etc.; medium risk conditions, such as 7 days within the same drug prescription 3 times, high price settlement of non-emergency prescriptions at night, drug store credit rating ≤ B level, etc.; low risk conditions, such as first prescription of chronic disease drugs, price fluctuation within 10%, drug store credit rating ≥ A level, etc. Any one of the high-risk conditions is determined as high-level risk, two or more medium-risk conditions are determined as medium-level risk, and only one low-risk condition or no risk is determined as low-level risk.
[0042] By acquiring multi-source historical data and training the model, a reliable model foundation is provided for subsequent intelligent pharmacy recommendation and prescription risk identification. The patient and pharmacy historical data are processed, such as outlier processing, missing value filling, feature enhancement, etc., to improve data quality, so that the trained pharmacy recommendation model can more accurately match patients and pharmacies; the risk identification model can more scientifically identify prescription risks by dividing risk levels and setting corresponding conditions, providing a basis for subsequent risk interception.
[0043] Second, prescription generation and evidence anchoring: doctors issue patient prescriptions through the HIS system, calculate the prescription hash value using the SHA-256 algorithm, and write the prescription hash value into the blockchain. After the doctor issues the patient's prescription, the structured prescription data is standardized, the prescription hash value is calculated, and the prescription hash value is written into the blockchain as the initial evidence anchor point. After the doctor issues the prescription, the hash value is calculated and written into the blockchain, using the characteristics of the blockchain that cannot be tampered with, the prescription hash value is used as the initial evidence anchor point, ensuring the originality and integrity of the prescription information, preventing the prescription from being maliciously tampered with, and providing a trusted basis for subsequent prescription verification and circulation.
[0044] Third step, intelligent pharmacy recommendation and prescription risk interception: Obtain patient historical data such as patient past selection preferences, price sensitivity, etc., pharmacy historical data such as compliance timeliness, complaint rate, credit rating, etc., and regulatory historical data such as violation records, drug quality incidents, etc. The model makes intelligent pharmacy recommendations based on current prescription information through three-party historical data, and identifies and processes current prescriptions at different levels based on prescription data and patient data. Real-time features of the current prescription are input into the intelligent pharmacy recommendation model, and the model generates pharmacy recommendation scores based on the real-time features of the current prescription and the historical data of patients, pharmacies, and regulators. The pharmacies are ranked from high to low based on the pharmacy recommendation scores for the patient to choose. Intelligent pharmacy recommendation based on three-party historical data and current prescription information can accurately recommend suitable pharmacies for patients, saving patients time and effort in selecting pharmacies and improving drug purchasing efficiency. The prescription is processed at different levels, high-risk prescriptions are immediately frozen and notified for verification, medium-risk prescriptions are manually reviewed, and low-risk prescriptions are monitored in real time. This can identify and handle prescriptions of different risk levels in a timely manner, effectively reduce drug risks, and protect patient safety; At the same time, record the abnormal prescription flow path and update the identification rules to continuously improve the risk identification capability.
[0045] Based on prescription data and patient data, the model identifies and processes current prescriptions at different levels. High-risk prescriptions are immediately frozen and the prescription flow is stopped. A freeze report is sent to multiple parties, and the medical insurance verification is notified to intervene. Medium-risk prescriptions are marked as pending and manually reviewed. Based on the review results, the prescription is either continued, upgraded to high-risk, or additional materials are provided. Prescriptions that are not reviewed within a certain time period are automatically upgraded to high-risk and a warning is issued. Low-risk prescriptions remain in the flow and are monitored in real time. The complete flow path of abnormal state prescriptions is recorded. Risk prescription feature data is extracted to determine if it is a new pattern. If it is a new pattern, a new identification rule is created and transmitted to the risk identification model database and pushed to all nodes. Otherwise, the existing identification rules are strengthened.
[0046] Fourth step, secure authorization and password sharing: Patients perform face recognition through the medical insurance platform and select a pharmacy to share the prescription from the list of intelligent recommendations. A one-time temporary sharing password, such as 6T9H-2K4P, is generated with a validity period of 10 minutes. Only the pharmacy selected by the patient can obtain the prescription within 10 minutes using the password. After the pharmacy uses the password, the password becomes invalid and cannot be used again. Patients perform secure identification through the medical insurance platform and generate a temporary sharing password. Only authorized pharmacies can obtain the prescription using the password, reducing the risk of prescription information leakage during sharing, protecting the privacy and security of patients' personal information and prescription information, and ensuring the safety and controllability of prescription sharing.
[0047] The fifth step is password identification and payment settlement: the pharmacy identifies the patient's temporary shared password to obtain the patient's prescription, verifies the patient's prescription related information, and performs medical insurance settlement after verification. After verification, the settlement through the blockchain is stored. When the pharmacy verifies the prescription, the hash value of the obtained prescription is verified with the hash value of the prescription stored in the blockchain. If the two hash values are consistent, the verification is passed, otherwise, the prescription is tampered with and an alarm is triggered. When the medical insurance settlement is completed, the payment voucher data is constructed, and the hash value of the voucher data obtained by using the SHA-256 algorithm is stored in the blockchain for medical insurance fund audit and pharmacy settlement to verify whether the voucher data hash value is consistent. If consistent, it means that the medical insurance settlement is compliant, otherwise, it means that the medical insurance settlement is illegal, triggering a financial audit alarm. The pharmacy obtains the prescription through the password and verifies the hash value, ensuring the authenticity and integrity of the prescription, and ensuring the accuracy of the prescription flow. After the medical insurance settlement, the payment voucher data hash value is stored in the blockchain for verification, ensuring the compliance of the medical insurance settlement, preventing the settlement data from being tampered with, facilitating the medical insurance fund audit and pharmacy settlement verification, and ensuring the safety of the medical insurance fund.
[0048] The sixth step is drug delivery and quality storage: obtain the related data information of the drug to be delivered to generate a drug delivery voucher, and upload the delivery voucher to the blockchain. Obtain the drug supervision code and temperature sensor data of the cold chain drug when the drug is delivered, construct the drug delivery voucher through the above data, calculate the drug itself hash value and environment hash value using the SHA-256 algorithm, combine the drug itself hash value and environment hash value to obtain the drug delivery hash value, and the patient receives the drug and scans to obtain the delivery verification hash value and compares it with the drug delivery hash value. If consistent, it means that the drug does not suspect to be replaced or illegally stored, otherwise, it means that the drug is suspected to be replaced or illegally stored. Obtain the drug related data to generate a delivery voucher and upload it to the blockchain, and the patient can confirm that the drug has not been replaced or illegally stored by comparing the hash values, ensuring the quality of the drug and ensuring the safety of the patient's medication. The blockchain storage realizes the traceability of the drug delivery information, provides a reliable basis for the traceability and responsibility identification of the drug quality problem. Through the prescription content hash, payment voucher hash and drug traceability hash, a three-chain verification mechanism is realized to build a complete evidence chain.
[0049] The seventh step is model retraining: after processing the related data of the current external prescription, input it into the model for retraining and optimization. The related data of the current external prescription is used for model retraining, which continuously optimizes the pharmacy recommendation and risk identification model, so that the model can better adapt to the changes of the actual business scene, improve the recommendation accuracy and risk identification ability of the model, and continuously improve the efficiency and reliability of the entire medical insurance external prescription flow and data supervision process.
[0050] The above embodiments only express several implementation manners of the present application, and the description is more specific and detailed, but it should not be understood as a limitation on the patent scope of the present application. It should be noted that for ordinary skilled persons in the art, without departing from the concept of the present application, several modifications and improvements can be made, which are all within the protection scope of the present application. Therefore, the protection scope of the patent of the present application should be subject to the appended claims.
Claims
1. A method for circulation and data supervision of medical insurance prescriptions, characterized in that: The specific steps include: S1. Build a pharmacy recommendation model and risk identification model: Obtain multi-source historical data, process the acquired historical data, and input it into the pharmacy recommendation model and risk identification model to train the model; S2. Prescription generation and evidence anchoring: The doctor writes a prescription to the patient through the HIS system, calculates the prescription hash value, and writes the prescription hash value into the blockchain; S3. Smart Pharmacy Recommendations and Prescription Risk Interception: This model obtains historical data from patients, pharmacies, and regulators. Based on this data, the model makes smart pharmacy recommendations based on current prescription information. Based on both prescription and patient data, the model identifies and intercepts current prescriptions in a graded manner. S4. Secure authorization and password-based sharing: Patients are securely identified through the medical insurance platform, select a pharmacy to share their prescription with from a list of intelligently recommended pharmacies, and generate a temporary sharing password. S5. Password identification and payment settlement: The pharmacy identifies the patient and temporarily shares the password to obtain the patient's prescription, verifies the patient's prescription-related information, and then settles the medical insurance after verification, and stores the settlement evidence on the blockchain; S6. Drug delivery and quality certification: Obtain relevant data information of the drugs to be delivered to generate drug delivery certificates, and upload the delivery certificates to the blockchain; S7. Model retraining: The relevant data of the external prescription is processed and input into the model, and the model is retrained and optimized.
2. A method for circulation and data supervision of medical insurance prescriptions according to claim 1, characterized in that: In step S1, building a pharmacy recommendation model specifically includes the following steps: S11. Obtain patient historical data and pharmacy historical data; S12. Extract key features of the patient's historical data and key features of the pharmacy's historical data to obtain the patient's key feature historical data and the pharmacy's key feature historical data; S13. Perform outlier processing and missing value filling on the patient key feature historical data and the pharmacy key feature historical data, and perform feature enhancement on the data to obtain second feature historical data; S14. Feature classification is performed on the second feature historical data and time features are extracted to construct a training data set. The training data set is divided into a training set, a test set, and a validation set in proportion to train the pharmacy recommendation model.
3. A method for circulation and data supervision of medical insurance prescriptions according to claim 1, characterized in that: In step S1, when training the risk identification model, the risk levels are divided into high, medium and low levels, and high-risk conditions, medium risk conditions and low-risk conditions are set respectively; if any one of the high-risk conditions is met, it is determined to be a high-level risk; if two or more medium-risk conditions are met, it is determined to be a medium-level risk; if only one low-risk condition is met or there is no risk, it is determined to be a low-level risk.
4. A method for circulation and data supervision of medical insurance prescriptions according to claim 1, characterized in that: In step S2, after the doctor prescribes the patient's prescription, the structured prescription data is normalized, the prescription hash value is calculated, and the prescription hash value is written into the blockchain as the initial evidence anchor point.
5. The method for circulating and supervising medical insurance prescriptions according to claim 1, characterized in that: In step S3, the real-time features of the current prescription are obtained and input into the intelligent pharmacy recommendation model. The model generates a pharmacy recommendation score based on the historical data of the patient, the pharmacy, and the regulator, as well as the real-time features of the current prescription, and ranks the recommendations from high to low according to the pharmacy recommendation score for the patient to choose.
6. A method for circulation and data supervision of medical insurance prescriptions according to claim 1, characterized in that: In step S3, based on the prescription data and patient data, the current prescription is intercepted and processed in a graded manner through model identification. For high-risk prescriptions, the prescription circulation is immediately frozen, and the freezing report is transmitted to multiple parties. The medical insurance verification intervention is notified, and the medium-risk prescription is marked as pending. Manual review is conducted and the review conclusion of continuing the circulation, upgrading to high-level risk or supplementing materials is made according to the review results. Prescriptions that have timed out and have not been reviewed are automatically upgraded to high-level risks for early warning processing. Low-risk prescriptions are kept in circulation and monitored in real time, and the complete circulation path of abnormal prescriptions is recorded; the characteristic data of the risk prescription is extracted to determine whether it is a new model. If it is determined to be a new model, a new identification rule is created and transmitted to the risk identification model database and pushed to all nodes. Otherwise, the existing identification rule is strengthened.
7. A method for circulation and data supervision of medical insurance prescriptions according to claim 1, characterized in that: In step S5, when the pharmacy verifies the prescription, it verifies the hash value of the prescription and the hash value of the prescription stored in the blockchain. If the two hash values are consistent, the verification is passed. Otherwise, it indicates that the prescription has been tampered with and triggers an alarm.
8. The method for circulating and supervising medical insurance prescriptions according to claim 1, characterized in that: In step S5, when the medical insurance settlement is completed, the payment voucher data is constructed, and the calculated voucher data hash value is stored in the blockchain for verification of whether the voucher data hash value is consistent during the medical insurance fund audit and pharmacy settlement. If consistent, it means that the medical insurance settlement is compliant. Otherwise, it means that the medical insurance settlement is in violation of regulations, triggering a financial audit alarm.
9. The method for circulating and supervising medical insurance prescriptions according to claim 1, characterized in that: In step S6, the drug supervision code at the time of drug delivery and the temperature sensor data of the cold chain drug are obtained, the drug delivery certificate is constructed based on the above data, the drug's own hash value and the environmental hash value are calculated, and the drug's own hash value and the environmental hash value are combined to obtain the drug delivery hash value. After receiving the drug, the patient scans the delivery verification hash value and compares it with the drug delivery hash value. If they are consistent, it means that the drug is not suspected of being swapped or stored in violation of regulations. Otherwise, it means that the drug is suspected of being swapped or stored in violation of regulations.