Medical supply data intelligent tracing and risk early warning system based on block chain

Abstract

The invention discloses a blockchain-based intelligent tracing and risk early warning system for medicine supply data, and relates to the technical field of data processing, and the system comprises the steps: carrying out the circulation node extraction of a medicine supply chain, and obtaining a plurality of medicine supply participants; performing distributed block deployment to obtain a medicine supply block chain; acquiring a medicine database, and acquiring a medicine supply data set; performing classification identification and mapping encryption, and uploading to the medicine supply block chain to obtain a block chain distributed account book; obtaining an access smart contract, and obtaining target drug supply data; and performing risk response early warning on the target drug based on the target drug supply data. The technical problems that in the prior art, traditional Chinese medicine supply data tracing is difficult and inaccurate, and potential risks of medicine supply are difficult to find and accurately evaluate in advance are solved, and the technical effects that accurate and rapid tracing of target medicine supply data is achieved, the potential supply risks are found in time, and effective response and early warning are carried out are achieved.

Description

Technical Field

[0001] The present invention relates to the field of data processing technology, and in particular to a blockchain-based intelligent tracing and risk warning system for pharmaceutical supply data. Background Art

[0002] In the pharmaceutical industry, ensuring the safety, stability and traceability of drug supply has always been a critical issue. Traditional pharmaceutical supply management methods often rely on paper records and decentralized databases, which makes data easy to lose and tamper with, and the traceability process is cumbersome and inaccurate. At the same time, risk assessment in pharmaceutical supply usually relies on manual experience and simple data analysis methods, which makes it difficult to fully and timely discover potential risks, and it is also impossible to accurately assess and warn of risks.

[0003] Existing technologies have technical problems such as difficulty and inaccuracy in tracing medical supply data, and difficulty in detecting and accurately assessing potential risks in medical supply in advance. Summary of the invention

[0004] This application provides a blockchain-based intelligent traceability and risk warning system for medicine supply data, which is used to solve the technical problems of the existing technology that medicine supply data is difficult and inaccurate to trace, and it is difficult to discover and accurately assess potential risks of medicine supply in advance.

[0005] In view of the above problems, this application provides a blockchain-based intelligent traceability and risk warning system for pharmaceutical supply data.

[0006] This application provides a blockchain-based intelligent traceability and risk warning system for pharmaceutical supply data, the system comprising: A supply chain extraction module, the supply chain extraction module is used to extract the flow nodes of the pharmaceutical supply chain to obtain multiple pharmaceutical supply participants, the multiple pharmaceutical supply participants include raw material suppliers, drug manufacturers, pharmaceutical logistics providers and drug users; a pharmaceutical supply blockchain acquisition module, the pharmaceutical supply blockchain acquisition module is used to perform distributed block deployment according to the multiple pharmaceutical supply participants to obtain a pharmaceutical supply blockchain, each block node in the pharmaceutical supply blockchain corresponds to a pharmaceutical supply participant one by one; a pharmaceutical supply data set acquisition module, the pharmaceutical supply data set acquisition module is used to obtain a pharmaceutical database, encode each pharmaceutical in the pharmaceutical database, obtain a pharmaceutical identification code set, and collect and obtain pharmaceutical supply data of the multiple pharmaceutical supply participants a block chain distributed ledger acquisition module, the block chain distributed ledger acquisition module classifies and identifies the medicine supply data set based on the medicine identification code set and maps and encrypts it and uploads it to the medicine supply blockchain to obtain a block chain distributed ledger; a medicine supply data acquisition module, the medicine supply data acquisition module is used to obtain an access smart contract through the medicine supply blockchain, trace the supply data of the block chain distributed ledger through the to-be-retrieved medicine identification and the access smart contract, and obtain the target medicine supply data; a risk response warning module, the risk response warning module performs feature extraction and risk assessment based on the target medicine supply data to obtain medicine supply risk parameters, and performs risk response warning on the target medicine through the medicine supply risk parameters.

[0007] One or more technical solutions provided in this application have at least the following technical effects or advantages: The supply chain extraction module is used to extract the circulation nodes of the pharmaceutical supply chain and obtain multiple pharmaceutical supply participants; the pharmaceutical supply blockchain acquisition module is used to perform distributed block deployment according to the multiple pharmaceutical supply participants to obtain the pharmaceutical supply blockchain; the pharmaceutical supply data set acquisition module is used to obtain the pharmaceutical database, encode each pharmaceutical in the pharmaceutical database, obtain a pharmaceutical identification code set, and collect and obtain the pharmaceutical supply data set of the multiple pharmaceutical supply participants; the blockchain distributed ledger acquisition module classifies and identifies the pharmaceutical supply data set based on the pharmaceutical identification code set and maps and encrypts it and uploads it to the pharmaceutical supply blockchain to obtain a blockchain distributed ledger; the drug supply data acquisition module is used to obtain access to the smart contract through the pharmaceutical supply blockchain, trace the supply data, and obtain the target drug supply data; the risk response warning module performs feature extraction and risk assessment based on the target drug supply data, obtains drug supply risk parameters, and performs risk response warning. The technical effect of achieving accurate and rapid tracing of the target drug supply data, timely discovering potential supply risks, and conducting effective response warnings is achieved. BRIEF DESCRIPTION OF THE DRAWINGS

[0008] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0009] Figure 1 A schematic diagram of the structure of a blockchain-based intelligent traceability and risk warning system for pharmaceutical supply data is provided for the embodiment of the present application; Figure 2 A schematic diagram of the structure execution of a drug supply data acquisition module of a blockchain-based drug supply data intelligent traceability and risk warning system is provided for an embodiment of the present application.

[0010] Explanation of the accompanying drawings: supply chain extraction module 10, pharmaceutical supply blockchain acquisition module 20, pharmaceutical supply data set acquisition module 30, blockchain distributed ledger acquisition module 40, pharmaceutical supply data acquisition module 50, risk response warning module 60. DETAILED DESCRIPTION

[0011] This application provides a blockchain-based intelligent traceability and risk warning system for medicine supply data, which is used to solve the technical problems of the prior art that medicine supply data is difficult and inaccurate to trace, and it is difficult to discover and accurately assess the potential risks of medicine supply in advance. The technical scheme in the embodiment of this application will be clearly and completely described below in conjunction with the drawings in the embodiment of this application. Obviously, the described embodiment is only a part of the embodiment of this application, not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of this application.

[0012] Embodiment 1, as Figure 1 As shown, the present application provides a blockchain-based intelligent traceability and risk warning system for pharmaceutical supply data, the system comprising: The supply chain extraction module 10 is used to extract flow nodes from the pharmaceutical supply chain to obtain multiple pharmaceutical supply participants, including raw material suppliers, drug manufacturers, drug logistics providers, and drug users.

[0013] Specifically, the main task of the supply chain extraction module 10 is to accurately identify and extract the circulation nodes in the pharmaceutical supply chain. In the pharmaceutical field, the operation of the supply chain involves many links and roles. Through the operation of this module, each key part can be clearly sorted out. For raw material suppliers, this is the starting point of the entire supply chain. They provide various raw materials necessary for drug production. Drug manufacturers are responsible for converting raw materials into drugs with specific efficacy and quality standards. Drug logistics companies bear the important responsibility of safely, promptly and accurately transporting the produced drugs to various sales and use locations. Drug users, that is, patients and medical institutions who ultimately use drugs, are the terminals of the entire supply chain. Through the precise extraction of these circulation nodes and the comprehensive sorting of each participant, it provides an indispensable foundation and prerequisite for the entire system to realize intelligent traceability and risk warning of pharmaceutical supply data.

[0014] The medicine supply blockchain acquisition module 20 is used to perform distributed block deployment according to the multiple medicine supply participants to obtain a medicine supply blockchain, and each block node in the medicine supply blockchain corresponds one-to-one to a medicine supply participant.

[0015] Specifically, the pharmaceutical supply blockchain acquisition module 20 plays a key role in building the core technical architecture in the entire system. It deploys distributed blocks according to the multiple pharmaceutical supply participants determined by the previous supply chain extraction module 10. This means that an independent and unique corresponding block node is created for each participant. For example, the raw material supplier has its own exclusive block, and the drug manufacturer, drug logistics provider and drug user also have their own corresponding blocks, which are dedicated to storing and processing the relevant data and transaction records of the corresponding participants. Taking the block node of the raw material supplier as an example, it contains detailed information such as the type, source, quality inspection report, and supply contract of the raw material, while the block of the drug manufacturer covers the production process parameters, batch records, quality inspection results, etc. This one-to-one distributed block deployment method ensures the independence and security of the data. Each participant can only access and operate his own block, reducing the risk of data tampering or leakage. Secondly, the scalability and flexibility of the system are improved, and new participants can easily join and create corresponding blocks. Through this precise block deployment, a complete and efficient pharmaceutical supply blockchain is finally formed, which lays a solid technical foundation for the trusted storage, sharing and traceability of pharmaceutical supply data.

[0016] The medicine supply data set acquisition module 30 is used to acquire a medicine database, encode each medicine in the medicine database to obtain a medicine identification code set, and collect and acquire the medicine supply data sets of the multiple medicine supply participants.

[0017] Specifically, the medicine supply data set acquisition module 30 is responsible for data collection and collation in the entire system. It is committed to obtaining a medicine database, which contains a large amount of detailed information about various medicines. Then, each medicine in the database is coded and a unique identification code is assigned to each medicine, thereby forming a collection of medicine identification codes. These identification codes help to quickly and accurately identify and manage different medicines. At the same time, this module also undertakes the important task of collecting medicine supply data sets from multiple medicine supply participants. For raw material suppliers, the collected data include raw material procurement records, quality inspection reports, inventory information, etc. For drug manufacturers, the acquired data may cover production plans, process parameters, quality control data, etc. The data of drug logistics providers may include transportation routes, transportation time, storage conditions, etc. The data of drug users includes drug usage feedback, demand forecasts, etc. Module 30 provides a rich and comprehensive medicine supply data foundation for the system, provides strong support for subsequent analysis, tracing and risk warning, and realizes accurate and complete data, efficient tracing, and timely and effective risk warning. The blockchain distributed ledger acquisition module 40 classifies and identifies the pharmaceutical supply data set based on the pharmaceutical identification code set, maps and encrypts the data, and uploads the data to the pharmaceutical supply blockchain to obtain a blockchain distributed ledger.

[0018] Specifically, the blockchain distributed ledger acquisition module 40 performs detailed classification and identification work on the pharmaceutical supply data set based on the pharmaceutical identification code set, which means that the data is grouped and classified according to the identification codes of different pharmaceuticals, so that the relevant data of the same medicine can be concentrated together for subsequent management and query. After the classification and identification is completed, the data is mapped and encrypted. The encryption operation can protect the security and privacy of the data and prevent unauthorized access and tampering. Then, the classified, identified and encrypted data is uploaded to the pharmaceutical supply blockchain. In this way, a complete blockchain distributed ledger is formed on the blockchain. This ledger contains rich and accurate pharmaceutical supply data, and due to the characteristics of the blockchain, the data has the characteristics of being tamper-proof, traceable and distributed storage.

[0019] The drug supply data acquisition module 50 is used to obtain the access smart contract through the medicine supply blockchain, trace the supply data of the blockchain distributed ledger through the drug identifier to be retrieved and the access smart contract, and obtain the target drug supply data.

[0020] Specifically, the drug supply data acquisition module 50 obtains access to the smart contract through the pharmaceutical supply blockchain. This smart contract stipulates the rules and permissions for data access to ensure that only legitimate requests can be executed. Then, the user enters the identifier of the drug to be retrieved, such as the name, batch number or specific code of the drug. The module uses this identifier of the drug to be retrieved and the access smart contract obtained to conduct in-depth supply data tracing on the blockchain distributed ledger. The blockchain distributed ledger contains detailed data from the procurement of raw materials, production and processing, logistics and transportation to final sales and use of drugs. Through precise tracing and screening, the supply data of the target drug is finally obtained. These data include a series of key information such as the production time, production location, source of raw materials, transportation route, storage conditions, etc. of the drug. Use this module to quickly obtain detailed supply data of problematic drugs, accurately determine the affected scope and links, and take effective recall measures in a timely manner.

[0021] The risk response warning module 60 performs feature extraction and risk assessment based on the target drug supply data to obtain drug supply risk parameters, and performs risk response warning on the target drug through the drug supply risk parameters.

[0022] Specifically, the risk response warning module 60 performs in-depth analysis based on the acquired target drug supply data, and extracts key feature information from complex data through feature extraction technology, such as abnormalities in the drug production cycle, delays in logistics and transportation, and fluctuations in raw material quality. The deep learning network structure is used to classify and identify this abnormal data set. Through training, two important parts are obtained: the supply risk type assessment semi-network and the supply risk level assessment semi-network. The supply risk type assessment semi-network can determine the specific type of risk, such as problems in the production link, delays in the logistics link, or unqualified raw material quality. The supply risk level assessment semi-network is responsible for assessing the severity of the risk, whether it is a mild, moderate or severe risk. Next, the two semi-networks are connected and merged to build a complete pharmaceutical supply risk assessment network. Finally, based on this built pharmaceutical supply risk assessment network, the previously extracted cluster abnormal feature data is subjected to an in-depth risk factor assessment, thereby obtaining the pharmaceutical supply risk parameters. Timely risk response warnings are provided for target drugs, and the warning methods include sending alarm information, prompting to take emergency measures, adjusting supply chain strategies, etc.

[0023] In a possible implementation, the medicine supply data set acquisition module 30 includes: Acquire pharmaceutical attribute factor information, fill the pharmaceutical attribute factor information based on the pharmaceutical database, and obtain a pharmaceutical attribute content set; use the pharmaceutical attribute content set as a tree node to construct a pharmaceutical attribute classification tree; use each tree node in the pharmaceutical attribute classification tree as a coding element, design an attribute coding rule, and the attribute coding rule includes element sequence, coding length, and coding identifier; perform attribute coding on each pharmaceutical in the pharmaceutical database based on the attribute coding rule to obtain the pharmaceutical identification code set.

[0024] Specifically, first obtain the various attribute factor information of pharmaceuticals, such as the name of the drug, who is the manufacturer, the dosage form of the drug (such as tablets, capsules, injections, etc.), specification size (content, dosage, etc.) and production date and other key information, with the pharmaceutical database as the basis and reference. This database stores a large amount of detailed specifications and standard information about various types of pharmaceuticals. Use this information in the database to supplement and improve the acquired pharmaceutical attribute factor information. For example, if the acquired pharmaceutical attribute factor information only contains the drug name and manufacturer, by searching for matching records in the pharmaceutical database, supplement the dosage form, specification and production date of the drug and other unobtained information. Finally, a complete and rich collection of pharmaceutical attribute content is formed, covering detailed information on many important aspects of the drug.

[0025] The previously acquired and filled pharmaceutical attribute content set is regarded as an independent unit, and each unit is regarded as a tree node. When constructing the pharmaceutical attribute classification tree, these nodes will be organized according to a certain logic and hierarchy. For example, the major categories of medicines are first used as the first-level branches, such as antibiotics, cardiovascular, digestive system, etc. Under each major category, the name of the drug is used as the second-level branch. For each specific drug, its manufacturer, dosage form, specification, production date and other attributes are expanded in sequence as lower-level branches. In this way, a hierarchical and clearly structured pharmaceutical attribute classification tree is formed.

[0026] In the drug attribute classification tree, each tree node represents specific drug attribute information, such as drug name, manufacturer, dosage form, specification, production date, etc. At the same time, these nodes are at different levels, reflecting the classification level of the attributes. Using these tree nodes as coding elements means constructing the code according to the specific content represented by the node and its position (i.e., level) in the classification tree. In terms of element order, it is necessary to determine the order in which each attribute is arranged in the code. For example, the drug name can be coded first, because it is the most direct key factor to distinguish different drugs; followed by the manufacturer, because it can reflect the production source; then the dosage form, specification, production date, etc. The setting of the code length should be determined according to the actual situation and needs. For the drug name, due to the diversity and complexity of the drug name, a longer code length is assigned, such as 10 digits; the manufacturer's code length can be relatively short, such as 6 digits, the dosage form may be 3 digits, the specification 5 digits, and the production date 8 digits. The coding identifier is a specific symbol or character combination used to clearly distinguish different attributes. For example, "AA" is used to identify the drug name, "BB" is used to identify the manufacturer, "CC" is used to identify the dosage form, "DD" is used to identify the specification, and "EE" is used to identify the production date. The attribute coding rules designed in this way have clear structures and meanings, and can provide efficient, accurate and consistent coding methods when processing and managing pharmaceutical data, facilitating data storage, retrieval, analysis and communication.

[0027] After the attribute coding rules are determined, each drug in the drug database will be coded according to these rules. According to the previously set element order, the various attributes of the drug are coded in turn, and the coding is strictly followed when coding. By performing such coding processing on all drugs in the database, a set of drug identification codes is finally obtained. Each identification code in this set uniquely corresponds to the detailed attributes of a drug. This coding method helps to quickly and accurately identify and manage drugs and improve the efficiency of data processing and retrieval.

[0028] In a possible implementation, the blockchain distributed ledger acquisition module 40 further includes: Based on the pharmaceutical identification code set, the pharmaceutical supply data set is classified, identified and integrated to obtain a pharmaceutical supply identification data set; the pharmaceutical database is encrypted by using the pharmaceutical identification code set to determine a pharmaceutical identification encryption coefficient set; according to the pharmaceutical identification encryption coefficient set, an encryption algorithm set is configured; based on the encryption algorithm set, the pharmaceutical supply identification data set is mapped, encrypted and uploaded to the pharmaceutical supply blockchain to obtain the blockchain distributed ledger.

[0029] Specifically, the pharmaceutical identification code set is a set of codes used to uniquely identify different pharmaceuticals, and the pharmaceutical supply data set contains a large amount of information related to pharmaceutical supply, such as the inventory quantity, transportation route, sales records, etc. of the drugs. When operating, the pharmaceutical identification code will be used as the basis for classification. For each piece of data in the pharmaceutical supply data set, the identification code is used to determine the pharmaceutical category to which it belongs. For example, pharmaceutical-related data with the identification code "001" will be classified into one category, and those with the identification code "002" will be classified into another category, and so on. After the classification is completed, the data of the same category will be integrated. This includes merging, sorting or aggregating multiple data under the same identification code according to specific rules. Through such a classification identification and integration process, the pharmaceutical supply identification data set is finally obtained. In this data set, each type of data clearly corresponds to a specific pharmaceutical identification code, which facilitates the subsequent independent analysis, processing and management of the supply of different pharmaceuticals.

[0030] The pharmaceutical identification code set provides a unique identification for each pharmaceutical. The pharmaceutical database stores a large amount of detailed information about these drugs, including the composition, production process, sales data, etc. When conducting encryption analysis, multiple factors are considered comprehensively. The first is the type of drug. For example, some special drugs, such as narcotic drugs or psychotropic drugs, may require a higher degree of encryption for their related data. Secondly, the sensitivity of production data also plays a key role. If the production data involves patented technology, core processes, or key links that may affect the quality and safety of drugs, these data will be considered highly sensitive and require stronger encryption protection. Different pharmaceuticals are evaluated and graded according to their type and the sensitivity of production data. Then, a specific encryption coefficient is determined for each pharmaceutical corresponding to the identification code, which reflects the encryption strength required for the drug data. Finally, a set of pharmaceutical identification encryption coefficients is formed, in which each coefficient corresponds to a specific pharmaceutical identification code, providing clear guidance for subsequent encryption operations.

[0031] The drug identification encryption coefficient set contains specific encryption coefficients for each drug. These coefficients reflect the encryption strength and complexity required for each drug data. When configuring the encryption algorithm set based on this set, certain rules and principles will be followed. If the encryption coefficient is high, it means that the corresponding drug data requires stronger confidentiality and security, and a complex and highly secure encryption algorithm will be selected, such as the 256-bit key version of the Advanced Encryption Standard (AES), or the Elliptic Curve Cryptography (ECC). If the encryption coefficient is low, it means that the sensitivity of the data is relatively weak, and a relatively simple but still effective encryption algorithm will be selected, such as the Data Encryption Standard (DES) or a hash function (such as SHA-256). By carefully matching the encryption coefficient and the characteristics of the encryption algorithm, each drug is configured with the encryption algorithm that best suits its confidentiality needs to form a complete encryption algorithm set.

[0032] There is already a previously configured encryption algorithm set. In this set, the identification data of each pharmaceutical product is matched with a corresponding appropriate encryption algorithm. For the pharmaceutical supply identification data set, this data set contains various pharmaceutical supply-related information that has been classified and identified. Using the encryption algorithm set, each part of the data in the data set is mapped and encrypted. The so-called mapping encryption is to encrypt the data using the encryption algorithm specified for the part of the data according to the predetermined correspondence. After the encryption is completed, the encrypted pharmaceutical supply identification data is uploaded to the pharmaceutical supply blockchain. Due to the characteristics of the blockchain, the uploaded data will be distributed and stored on multiple nodes, and has the advantages of being tamper-proof and traceable. Finally, a blockchain distributed ledger is formed on the blockchain, which contains the encrypted pharmaceutical supply identification data, ensuring the security and integrity of the data, and also facilitating accurate and reliable query and traceability when needed.

[0033] In a possible implementation, the blockchain distributed ledger acquisition module 40 further includes: According to the information on sensitive factors of pharmaceuticals, a pharmaceutical encryption evaluation coordinate system is constructed; based on the information on sensitive factors of pharmaceuticals, a sensitivity evaluation is performed on each pharmaceutical in the pharmaceutical database to generate a set of pharmaceutical sensitivity vectors; the modulus of the set of pharmaceutical sensitivity vectors is used as a set of encryption evaluation coefficients; the set of encryption evaluation coefficients is associated and mapped with the set of pharmaceutical identification codes to determine the set of pharmaceutical identification encryption coefficients.

[0034] Specifically, the information on sensitive factors of pharmaceuticals usually covers multiple aspects, such as the specificity of the ingredients of the drugs, the importance and urgency of the drugs' treatment of the diseases, the cost and technical difficulty of drug research and development, the complexity of the drug production process, and the scarcity of drugs in the market. When constructing the coordinate system for encrypted evaluation of pharmaceuticals, the key elements of these sensitive factors will be selected as the coordinate axes. The three factors of specificity of drug ingredients, importance of treating diseases, and market scarcity can be selected. On the axis of specificity of drug ingredients, one end can represent ordinary and common ingredients, and the other end can represent rare, unique ingredients or highly confidential formulas. One end of the axis of importance of treating diseases represents the treatment of general diseases, and the other end represents the treatment of major diseases that endanger life or seriously affect the quality of life. One end of the axis of market scarcity is sufficient market supply, and the other end is extreme shortage of the market. Through this setting, each pharmaceutical can have a corresponding position in this coordinate system, so that its sensitivity can be intuitively evaluated.

[0035] Information on sensitive factors of pharmaceuticals usually covers many aspects, such as the uniqueness of the formula of the drug, the degree of efficacy of the drug, the severity of the side effects of the drug, the intensity of market competition for the drug, the strictness of drug supervision, etc. When evaluating the sensitivity of each drug in the pharmaceutical database, clear evaluation criteria and score ranges will be set for each sensitive factor. Taking the uniqueness of the drug formula as an example, if the drug formula is common and easy to imitate, it may be given 1-3 points; if it is a unique formula with patent protection and extremely difficult to imitate, it will be given 7-10 points. For the degree of efficacy of the drug, if the drug can cure terminal illness or have a significant improvement effect on serious diseases, it will be given 8-10 points; for common diseases, it will be given 3-7 points. Each drug is evaluated and scored one by one on each sensitive factor. Suppose an innovative anti-cancer drug has a unique formula and a patent, and its efficacy is extremely significant, but its side effects are also obvious. There are almost no competitors in the market and it is strictly regulated. 9 points are given for uniqueness of formula, 9 points for significant efficacy, 7 points for severity of side effects, 2 points for fierce market competition, and 9 points for strict supervision. These scores are combined to form a vector, such as (9, 9, 7, 2, 9), which is the sensitivity vector of the drug. All drugs in the drug database are evaluated and scored in this way to generate a series of sensitivity vectors, which together constitute the drug sensitivity vector set.

[0036] For each sensitivity vector in the set of pharmaceutical sensitivity vectors, its modulus needs to be calculated. For example, for the sensitivity vector (3, 4, 5), its modulus is √(3²+4²+5²)=√50. Such a modulus calculation is performed on each vector in the set of pharmaceutical sensitivity vectors, and the values ​​of these moduli constitute the encryption evaluation coefficient set. These modulus values ​​reflect the comprehensive strength of the sensitivity of the corresponding pharmaceuticals. The larger the modulus value, the more sensitive the comprehensive performance of the pharmaceutical in various sensitive factors is, and a higher strength encryption is required; the smaller the modulus value, the lower the sensitivity is, and the encryption strength can be appropriately reduced.

[0037] The encryption assessment coefficient set contains a series of numerical values, each of which represents a comprehensive assessment result of the sensitivity of a pharmaceutical. The pharmaceutical identification code set is a unique identifier assigned to each pharmaceutical. When performing association mapping, a one-to-one correspondence is established between each encryption assessment coefficient and the corresponding pharmaceutical identification code. Through such an association, the pharmaceutical identification encryption coefficient set is determined, which means that when it is necessary to encrypt the data of a specific pharmaceutical, the corresponding encryption coefficient can be quickly found through its identification code, so as to clarify what strength of encryption measures should be adopted.

[0038] In a possible implementation, the blockchain distributed ledger acquisition module 40 further includes: The access smart contract is called through the medicine supply blockchain, and the permission access scope of the medicine supply data set is obtained according to the access smart contract; the medicine supply data set is preprocessed based on the permission access scope to obtain medicine supply data parameters; zero-knowledge proof technology is used to generate a zero-knowledge data range proof for the permission access scope and the medicine supply data parameters, and the zero-knowledge data range proof is uploaded to the blockchain distributed ledger.

[0039] Specifically, there is an access smart contract in the pharmaceutical supply blockchain. When it is necessary to determine the scope of access to the pharmaceutical supply data set, this smart contract is called through the blockchain mechanism. Role-based access control rules are defined in the access smart contract, and different roles are given different access rights. For example, the "drug regulatory role" is given higher permissions due to its responsibilities and needs, and can trace the drug supply data on the entire chain, which means that they can view the supply data of the entire chain from drug production, transportation, storage to sales. Other roles, such as ordinary drug suppliers, can only access data related to their own supply links, and drug distributors can only see part of the data related to the distribution link. Through such clear and detailed permission settings, the security and privacy of data can be ensured, while allowing participants with appropriate permissions to obtain the information they need to effectively perform their duties and make relevant decisions.

[0040] Once the scope of access permissions is determined, the pharmaceutical supply dataset is preprocessed. This means carefully checking each field and each record in the dataset. First, the specific access permissions for each field in the dataset are determined. For example, for the price field, only specific personnel or senior managers in the purchasing department can access it; the inventory field is only open to warehouse managers and production planning departments. Then, based on the settings of these access permissions, the data that meets the current access scope is filtered and extracted from the dataset. For example, if the current user's role can only access the name and basic description of the drug, then during the preprocessing process, only the relevant data of these fields will be obtained from the dataset, and sensitive information such as price and inventory will be excluded. Through such preprocessing, the final result is the pharmaceutical supply data parameters that meet the current permissions. These parameters can be safely provided to users with corresponding permissions, while avoiding the improper leakage of sensitive information.

[0041] Zero-knowledge proof technology is a cryptographic method that can prove to the verifier that something is true without revealing specific information. In this case, the access scope and the pre-processed pharmaceutical supply data parameters are input into the zero-knowledge proof algorithm. This algorithm will generate a zero-knowledge data range proof, which can show that the data parameters provided are indeed within the specified access scope, but will not disclose the specific access scope and the content of the data parameters. For example, assuming that the access scope stipulates that only the supply data of a certain number of drugs can be accessed, and the data parameters actually obtained are the relevant information of these drugs, the zero-knowledge proof will prove that the data provided meets the permission requirements without revealing which drugs and the specific data content. After the proof is generated, it is uploaded to the blockchain distributed ledger. In this way, other nodes on the blockchain can verify the validity of the proof, while being unable to obtain specific sensitive information, thereby achieving the credibility and traceability of data access under the premise of ensuring data privacy and security.

[0042] In one possible implementation, Figure 2 As shown, the drug supply data acquisition module 50 also includes: The access smart contract is called to download the zero-knowledge data range proof from the blockchain distributed ledger, and the range access data is obtained according to the zero-knowledge data range proof; the hash value of the range access data is compared and judged with the hash value of the medicine supply data set to obtain the supply data integrity result; when the supply data integrity results are consistent, the range access data is decrypted and traced based on the encryption algorithm set to obtain the target medicine supply data, and the tracing behavior data is recorded on the chain.

[0043] Specifically, when specific data access rights and related data are required, the access smart contract will be activated. By calling this smart contract, it will find and download the previously uploaded zero-knowledge data range proof from the blockchain distributed ledger. This proof is an important credential for confirming the scope and legitimacy of data access. With the zero-knowledge data range proof, the system will determine the scope of access data that can be obtained based on the information contained therein. For example, if the proof shows that a user or role has the right to access the supply data of a specific type of drug within a specific time period, the system will filter and extract data that meets these conditions from the entire pharmaceutical supply data set.

[0044] The hash value is a fixed-length string obtained by calculating the data through a specific hash function. It is unique and deterministic. Even if the data has a slight change, its hash value will be completely different. For the range access data and the entire medical supply data set, their hash values ​​are calculated respectively, and then the hash value of the range access data is compared with the hash value of the medical supply data set. If the two are exactly the same, it means that the range access data has not been changed in the process of being extracted from the data set, and the data is complete, thereby obtaining a complete supply data integrity result. On the contrary, if the two hash values ​​are different, it means that the range access data is inconsistent with the original medical supply data set in some aspects, and there is a situation where the data has been tampered with, lost, or extracted incorrectly, resulting in an incomplete data result.

[0045] When the supply data integrity results are consistent, it means that the range access data has not been tampered with and is complete and reliable. Next, these range access data are decrypted based on the previously configured encryption algorithm set. Because these data are encrypted during storage and transmission, to ensure data security. Through the correct decryption operation, the target drug supply data can be traced and obtained, that is, the required detailed, accurate and unencrypted drug supply related information can be obtained. At the same time, the relevant data of the traceability behavior will be recorded on the chain, including the identity of the tracer (such as user name, identity, etc.), the specific content of the data traced (such as the supply data of which drug), the time of tracing (accurate to the specific date and time point), etc. It is of great significance to record these traceability behavior data on the chain. On the one hand, it can ensure the traceability and non-tamperability of the traceability behavior, which is convenient for subsequent auditing and supervision; on the other hand, it also helps to establish a transparent and reliable data traceability system.

[0046] In a possible implementation, the risk response warning module 60 further includes: A target for drug supply anomaly analysis is obtained, and based on the target for drug supply anomaly analysis, feature extraction is performed on the target drug supply data to obtain a drug supply feature data set; cluster analysis is performed on the drug supply feature data sets to obtain feature cluster analysis results, and based on the feature cluster analysis results, abnormal feature data within the cluster is determined; risk factor assessment is performed based on the abnormal feature data within the cluster to obtain drug supply risk parameters, wherein the drug supply risk parameters include supply risk type and supply risk level.

[0047] Specifically, we first need to clarify the specific goals of the analysis of drug supply anomalies, including adverse drug reactions, price anomalies, purchase volume anomalies, and ingredient safety assessment. If the goal is to analyze adverse drug reactions, the features extracted from the target drug supply data include patient symptom reports after using the drug, recovery status after taking the drug, etc. If the goal is price anomalies, the extracted features cover drug prices in different time periods, market average price comparisons, price fluctuation ranges and frequencies, etc. For purchase volume anomalies, the extracted features include the number of purchase orders, the trend of changes within the purchase cycle, and the comparison with other drug purchase volumes during the same period. If it is for ingredient safety assessment, the features include the composition of the drug, the content ratio of each ingredient, the source and quality of the ingredients, etc. By extracting features for these specific goals, we can finally obtain a set of drug supply feature data.

[0048] The drug supply feature data set contains a variety of features related to drug supply, such as drug prices, purchase volume, inventory levels, transportation time, etc. When performing cluster analysis, select the K-Means algorithm. First, determine how many clusters the data should be divided into. For example, set it to 5 clusters (K=5). The algorithm will randomly select 5 initial center points, and then assign each data point to the closest cluster based on the distance between the data point and these center points. After the assignment is completed, recalculate the center point of each cluster. This process is repeated until the center point no longer changes significantly or the preset number of iterations is reached. Suppose that 5 clusters are finally obtained: Cluster 1, Cluster 2, Cluster 3, Cluster 4, and Cluster 5. For Cluster 1, assume that it contains 20 data points, which represent the price characteristics of a certain drug. The average price of these 20 data points is calculated to be 15 yuan, and the standard deviation is 3 yuan. Set a threshold, such as 2 times the standard deviation, that is, 6 yuan. If the price of a data point is 8 yuan, and its difference from the average is 7 yuan, which is greater than 6 yuan, then this data point is identified as abnormal feature data in Cluster 1. The same method is used to analyze clusters 2, 3, and other clusters. Through such detailed and specific calculations and comparisons, data points that obviously deviate from the normal distribution within the cluster can be accurately found in each cluster, that is, abnormal characteristic data within the cluster.

[0049] After determining the abnormal characteristic data within the cluster, the next step is to evaluate the risk factors and analyze the possible risk types caused by different abnormal characteristic data. For example, if the abnormal characteristic is a sudden and sharp increase in drug prices, then the supply risk type is the risk of increased costs; if the purchase volume suddenly drops sharply, the risk type is the risk of supply shortages. Then, the severity of these risks is evaluated to determine the supply risk level, which can usually be divided into high, medium, and low. For example, in the case of price increases, if the increase exceeds 50%, it is assessed as a high risk level; if the increase is between 20%-50%, it is a medium risk level; and if the increase is below 20%, it is a low risk level. For another example, a decrease in purchase volume of more than 80% is considered a high risk level, a decrease of 50%-80% is considered a medium risk level, and a decrease of less than 50% is considered a low risk level. Through such an evaluation, the drug supply risk parameters including supply risk types (such as cost increase risk, supply shortage risk, etc.) and supply risk levels (high, medium, and low) are finally obtained.

[0050] In a possible implementation, the risk response warning module 60 further includes: A pharmaceutical supply anomaly data set is collected and acquired, and classification and identification training is performed on the pharmaceutical supply anomaly data set using a deep learning network structure to obtain a supply risk type assessment semi-network and a supply risk level assessment semi-network; the supply risk type assessment semi-network and the supply risk level assessment semi-network are connected and merged to build a pharmaceutical supply risk assessment network; risk factors of the abnormal feature data within the cluster are assessed based on the pharmaceutical supply risk assessment network to obtain the pharmaceutical supply risk parameters.

[0051] Specifically, data related to various pharmaceutical supply anomalies are collected extensively. The sources of these data include internal databases of pharmaceutical companies, industry reports, records of regulatory authorities, etc. These data should cover a variety of different abnormal situations, such as sudden large-scale shortages of certain drugs, sharp price fluctuations in a short period of time, and serious quality problems of drugs leading to recalls. Then, a suitable convolutional neural network (CNN) is selected to input the collected pharmaceutical supply anomaly data set into this deep learning network for training. During the training process, classification labels are added to the data, for example, different supply anomalies are marked as different supply risk types (such as inventory risk, quality risk, market demand risk, etc.), and corresponding supply risk levels (high, medium, and low). The network will automatically learn the relationship between the features in the data and these classification labels. After multiple iterations of training and optimization, the network can gradually accurately identify and distinguish different risk types and levels. Finally, the network will be divided into two parts, one part is dedicated to evaluating the supply risk type, forming a supply risk type evaluation semi-network; the other part focuses on evaluating the supply risk level, becoming a supply risk level evaluation semi-network. For example, in the training data, when the inventory of medicines is below the safety stock line for three consecutive months, it is marked as inventory risk type, and when the inventory is more than 50% below the safety stock line, it is marked as high risk level. By learning these marked data, one part of the network can accurately determine whether the newly input data belongs to the inventory risk type, and the other part can determine whether its risk level is high, medium or low.

[0052] After obtaining the supply risk type assessment semi-network and the supply risk level assessment semi-network respectively, the next step is to connect and merge them to build a complete pharmaceutical supply risk assessment network. Determine the appropriate connection points between the two semi-networks. These connection points are usually certain nodes of the output or intermediate layer of the two semi-networks. Their characteristics and information can match and fuse with each other. Use the connection layer or shared hidden layer in the neural network to physically connect the two semi-networks at these connection points so that data and information can flow and interact between the two parts. When connecting, it is also necessary to consider the matching of the dimension, format and scale of the data. Some preprocessing and conversion operations may be required to ensure smooth and effective connection. After such careful connection and integration, a unified pharmaceutical supply risk assessment network that can simultaneously assess the supply risk type and level is finally formed.

[0053] With the established pharmaceutical supply risk assessment network, the previously determined intra-cluster abnormal feature data can be input into the network for risk assessment. These intra-cluster abnormal feature data contain various abnormal information related to pharmaceutical supply, such as abnormal fluctuations in the purchase volume of certain drugs, large changes in prices, and sharp rises and falls in inventory levels. When these data are input into the network, the network will analyze and process these data based on the knowledge and patterns it has learned. The neurons in the network will calculate and transmit the input data, and finally output the assessment results, i.e., the pharmaceutical supply risk parameters, through multiple layers of calculations and feature extraction. This output is presented in a specific format, such as an array containing risk type labels (such as inventory shortage risk, price fluctuation risk, etc.) and risk level values ​​(1 for high risk, 2 for medium risk, and 3 for low risk), which can timely detect potential supply risks and provide early warnings.

[0054] It should be noted that the above-mentioned sequence of the embodiments of the present application is only for description and does not represent the advantages and disadvantages of the embodiments. And the above-mentioned specific embodiments of this specification are described. In addition, the processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0055] The above description is only a preferred embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present application should be included in the protection scope of the present application.

[0056] This specification and the drawings are merely exemplary illustrations of the present application and are deemed to cover any and all modifications, variations, combinations or equivalents within the scope of the present application. Obviously, a person skilled in the art may make various modifications and variations to the present application without departing from the scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the present application and its equivalents, the present application intends to include these modifications and variations.

Claims

1. The blockchain-based intelligent traceability and risk warning system for pharmaceutical supply data is characterized by: The system comprises: A supply chain extraction module, which is used to extract flow nodes from the pharmaceutical supply chain to obtain multiple pharmaceutical supply participants, including raw material suppliers, pharmaceutical manufacturers, pharmaceutical logistics providers, and pharmaceutical users; A medicine supply blockchain acquisition module, the medicine supply blockchain acquisition module is used to perform distributed block deployment according to the multiple medicine supply participants to obtain a medicine supply blockchain, wherein each block node in the medicine supply blockchain corresponds to a medicine supply participant one by one; a medicine supply data set acquisition module, the medicine supply data set acquisition module is used to acquire a medicine database, encode each medicine in the medicine database to obtain a medicine identification code set, and collect and acquire medicine supply data sets of the plurality of medicine supply participants; A blockchain distributed ledger acquisition module, wherein the blockchain distributed ledger acquisition module classifies and identifies the pharmaceutical supply data set based on the pharmaceutical identification code set, maps and encrypts the data set, and uploads the data set to the pharmaceutical supply blockchain to obtain a blockchain distributed ledger; A drug supply data acquisition module, which is used to obtain access to the smart contract through the medicine supply blockchain, trace the supply data of the blockchain distributed ledger through the drug identifier to be retrieved and the access smart contract, and obtain the target drug supply data; A risk response warning module performs feature extraction and risk assessment based on the target drug supply data to obtain drug supply risk parameters, and performs risk response warning on the target drug through the drug supply risk parameters.

2. The blockchain-based intelligent traceability and risk warning system for pharmaceutical supply data according to claim 1, characterized in that: The step of obtaining a set of pharmaceutical identification codes includes: Acquire pharmaceutical attribute factor information, and fill the pharmaceutical attribute factor information based on the pharmaceutical database to obtain a pharmaceutical attribute content set; Using the pharmaceutical attribute content set as a tree node, constructing a pharmaceutical attribute classification tree; Taking each tree node in the pharmaceutical attribute classification tree as a coding element, designing an attribute coding rule, wherein the attribute coding rule includes element sequence, coding length and coding identifier; Attribute encoding is performed on each medicine in the medicine database based on the attribute encoding rule to obtain the medicine identification code set.

3. The blockchain-based intelligent traceability and risk warning system for pharmaceutical supply data as claimed in claim 1, characterized in that: The obtained blockchain distributed ledger includes: Classifying and identifying the medicine supply data set based on the medicine identification code set and integrating it to obtain a medicine supply identification data set; Performing encryption degree analysis on the pharmaceutical database through the pharmaceutical identification code set to determine a pharmaceutical identification encryption coefficient set; According to the encryption coefficient set of the pharmaceutical identification, an encryption algorithm set is configured; Based on the encryption algorithm set, the medicine supply identification data set is mapped, encrypted and uploaded to the medicine supply blockchain to obtain the blockchain distributed ledger.

4. The blockchain-based intelligent traceability and risk warning system for pharmaceutical supply data as claimed in claim 3 is characterized in that: The step of determining a set of encryption coefficients for drug identifications includes: According to the sensitive factor information of pharmaceuticals, an encrypted evaluation coordinate system for pharmaceuticals is constructed; Based on the drug sensitivity factor information, a sensitivity evaluation is performed on each drug in the drug database to generate a drug sensitivity vector set; Using the modulus of the pharmaceutical sensitivity vector set as a set of encrypted evaluation coefficients; The encrypted evaluation coefficient set is associated and mapped with the pharmaceutical identification code set to determine the pharmaceutical identification encryption coefficient set.

5. The blockchain-based intelligent traceability and risk warning system for pharmaceutical supply data as claimed in claim 3 is characterized in that: The system comprises: Invoking the access smart contract through the medicine supply blockchain, and obtaining the permission access scope of the medicine supply data set according to the access smart contract; preprocessing the medicine supply data set based on the permission access scope to obtain medicine supply data parameters; Use zero-knowledge proof technology to generate a zero-knowledge data range proof for the permission access scope and the medical supply data parameters, and upload the zero-knowledge data range proof to the blockchain distributed ledger.

6. The blockchain-based intelligent traceability and risk warning system for pharmaceutical supply data according to claim 5, characterized in that: The obtaining of target drug supply data includes: Calling the access smart contract to download the zero-knowledge data range proof from the blockchain distributed ledger, and obtaining range access data according to the zero-knowledge data range proof; Compare and judge the hash value of the range access data with the hash value of the medicine supply data set to obtain a supply data integrity result; When the supply data integrity results are consistent, the supply data decryption and tracing is performed on the range access data based on the encryption algorithm set to obtain the target drug supply data, and the tracing behavior data is recorded on the chain.

7. The blockchain-based intelligent traceability and risk warning system for pharmaceutical supply data according to claim 1, characterized in that: The drug supply risk parameters obtained include: Acquire a drug supply anomaly analysis target, perform feature extraction on the target drug supply data based on the drug supply anomaly analysis target, and obtain a drug supply feature data set; Performing cluster analysis on the drug supply feature data sets respectively to obtain feature cluster analysis results, and determining abnormal feature data within the cluster according to the feature cluster analysis results; A risk factor assessment is performed based on the abnormal characteristic data within the cluster to obtain a drug supply risk parameter, wherein the drug supply risk parameter includes a supply risk type and a supply risk level.

8. The blockchain-based intelligent traceability and risk warning system for pharmaceutical supply data according to claim 7, characterized in that: The drug supply risk parameters obtained include: Collecting and acquiring a pharmaceutical supply anomaly data set, and using a deep learning network structure to perform classification and identification training on the pharmaceutical supply anomaly data set to obtain a supply risk type assessment semi-network and a supply risk level assessment semi-network; Connect and merge the supply risk type assessment semi-network and the supply risk level assessment semi-network to build a medicine supply risk assessment network; Based on the medicine supply risk assessment network, risk factor assessment is performed on the abnormal feature data within the cluster to obtain the medicine supply risk parameter.

Images