Blockchain-based traditional Chinese medicine famous doctor experience prescription storage and sharing system

Abstract

The application discloses a blockchain-based traditional Chinese medicine famous doctor experience prescription storage and sharing system and belongs to the field of traditional Chinese medicine digitization and blockchain. The system solves the technical defects of the existing traditional Chinese medicine famous doctor experience prescription, such as unclear ownership definition, easy tampering of inheritance, insufficient privacy protection of sharing, and lack of credible data support for clinical transformation. The system combines the tamper-proof characteristics of blockchain and the structured analysis and image recognition capabilities of artificial intelligence to construct a whole-process system of famous doctor qualification verification, experience prescription structured analysis, multi-dimensional feature storage, blockchain hierarchical storage, intelligent sharing matching, and clinical feedback iteration. Through three-dimensional data encapsulation storage of 'text-physical- clinical', a public chain-permission chain hierarchical sharing architecture is realized to realize experience prescription ownership traceability, data tamper-proof, controllable sharing, and verifiable application. Through clinical feedback, an experience prescription iterative optimization closed loop is formed to provide a systematic technical solution for the inheritance, sharing and clinical transformation of traditional Chinese medicine famous doctor experience prescriptions.

Description

Technical Field

[0001] This invention relates to the field of digitalization of traditional Chinese medicine and blockchain, and in particular to a blockchain-based system for storing and sharing the experience prescriptions of renowned traditional Chinese medicine practitioners. Background Technology

[0002] The experienced prescriptions of renowned traditional Chinese medicine practitioners are the crystallization of clinical wisdom in TCM, embodying the long-term diagnostic and treatment practices and academic accumulation of these practitioners. They possess extremely high medical, scientific research, and clinical translational value. However, in the current process of inheriting and applying these experienced prescriptions, there are many systemic technical deficiencies that urgently need to be addressed:

[0003] The lack of credible evidence for defining ownership means that knowledge is often passed down orally or in writing, which can easily lead to ownership disputes during the digitization process. Furthermore, traditional storage methods cannot prevent data tampering, making it difficult to guarantee the authenticity and authority of knowledge providers.

[0004] The information on empirical prescriptions lacks standardized analysis. Core information such as diagnostic reasoning, compatibility ratios, and special decoction procedures are mostly vague descriptions. Information distortion is prone to occur during transcription and inheritance, leading to deviations in clinical application effects.

[0005] The evidence storage is limited to a single dimension, focusing only on textual information storage, without linking the physical characteristics of medicinal materials, typical clinical data, and empirical prescription texts for evidence storage, thus lacking multi-dimensional evidence for authenticity verification.

[0006] The imbalance between data sharing and privacy protection means that existing sharing models either make it difficult for practitioners to promote their expertise due to over-protection, or lead to the leakage of core expertise by renowned practitioners due to a lack of access control. Furthermore, the problem of data silos between medical institutions is serious, and practitioners' clinical application data cannot be effectively integrated.

[0007] The lack of a clinical feedback and iteration mechanism means that once empirical formulas are certified, they remain in a static state and cannot be dynamically optimized based on the actual effects of clinical application, resulting in low clinical translation efficiency.

[0008] In existing technologies, some solutions attempt to apply blockchain to the storage of TCM data or use artificial intelligence for prescription analysis, but none of them have formed a systematic solution that combines the special characteristics of the experience prescriptions of famous TCM doctors. Among the more typical published patents are: "A method and system for storing TCM prescriptions based on blockchain" with publication number CN115659528A, and "Intelligent sharing and matching system for TCM experience prescriptions" with publication number CN114282798A. In light of the technical advantages of this patent, both of the aforementioned published patents have significant shortcomings: Patent CN115659528A only achieves single-text-level on-chain storage of TCM prescriptions, lacking a dedicated mechanism for verifying the qualifications of renowned practitioners, thus failing to confirm the ownership of the prescriptions. Furthermore, it does not involve structured analysis and multi-dimensional feature verification of the prescriptions, making it difficult to guarantee the authenticity and authority of the stored data. It also lacks a clinical feedback iteration module, leaving the stored prescriptions in a static state, unable to achieve dynamic optimization. While patent CN114282798A involves sharing and matching prescriptions, it does not employ a hierarchical storage architecture, failing to balance the openness of sharing prescriptions with the privacy protection of core practitioner experiences. Moreover, the matching mechanism is based solely on basic symptom keywords, without incorporating TCM knowledge graphs and patient constitution characteristics, resulting in low matching accuracy. Additionally, it does not utilize blockchain technology for end-to-end traceability and lacks ownership protection functions, thus failing to protect the intellectual property rights of renowned TCM practitioners. In addition, existing solutions generally suffer from problems such as either only achieving single-dimensional data on-chaining without solving the issues of ownership confirmation and standardized analysis, or only focusing on prescription matching without a full-process evidence storage and feedback iteration mechanism, which cannot meet the actual needs of the inheritance and application of the experience prescriptions of famous traditional Chinese medicine practitioners.

[0009] Therefore, there is a need for an integrated system that can achieve credible evidence storage, ownership protection, intelligent sharing, clinical validation, and dynamic iteration. Summary of the Invention

[0010] Purpose of the invention: The invention provides a blockchain-based system for storing and sharing the experience prescriptions of renowned traditional Chinese medicine practitioners, which solves the problems mentioned in the background technology.

[0011] Technical Solution: To address the aforementioned technical problems, according to one aspect of the present invention, more specifically, a blockchain-based system for storing and sharing the experience prescriptions of renowned traditional Chinese medicine practitioners, this invention integrates the immutability of blockchain with the structured analysis and image recognition capabilities of artificial intelligence. It constructs a comprehensive system encompassing practitioner qualification verification, structured analysis of experience prescriptions, multi-dimensional feature storage, hierarchical blockchain storage, intelligent sharing and matching, and clinical feedback iteration. Through three-dimensional data encapsulation and storage ("text-physical-clinical") and a hierarchical sharing architecture (public blockchain-permissioned blockchain), it achieves traceable ownership of experience prescriptions, immutable data, controllable sharing, and verifiable applications. Specifically, it includes a practitioner qualification verification module, a structured analysis module for experience prescriptions, a multi-dimensional feature storage module, a hierarchical blockchain storage network, an intelligent sharing and matching module, and a clinical feedback iteration module. The functions and implementation logic of each module are as follows:

[0012] 1. Qualification Verification Module for Renowned Experts

[0013] This system is used to collect the identity information, professional qualification certificates, academic lineage records, and achievement certification materials of renowned traditional Chinese medicine practitioners. A multi-source cross-verification mechanism is used to confirm their qualifications and generate a practitioner's identity hash certificate containing a digital signature, which serves as the core basis for determining the ownership of the practitioner's credentials. The multi-source cross-verification mechanism includes: connecting to the National Administration of Traditional Chinese Medicine's database of practicing physicians and the certification system for traditional Chinese medicine academic organizations to obtain authoritative qualification verification data; using digital signature technology to digitize and store the practitioner's handwritten signatures and seal imprints, generating signature hash values; and conducting distributed verification through the master-disciple relationship chain nodes in the academic lineage, with each lineage node signing and confirming the qualification information to reach a consensus. The practitioner's identity hash certificate is generated by combining the qualification verification data hash, the signature hash value, and the verification signature from the lineage nodes. This certificate will serve as the core identifier for the practitioner's evidence storage, sharing, and ownership changes, and will be permanently bound to the practitioner's data.

[0014] 2. Empirical Square Structured Parsing Module

[0015] This method is used to obtain the original text, diagnostic reasoning, and key points of decoction techniques of traditional Chinese medicine masters' experience prescriptions. Through word segmentation, terminology alignment, and behavior instruction extraction, standardized experience prescription parameters are generated, which include the composition of medicinal materials, compatibility ratio, scope of application for diagnosis, special decoction operations, and contraindications. A hash algorithm is then used to generate a summary value of the standardized experience prescription parameters, which serves as the core anchor point for blockchain evidence storage. The processing flow of this module includes: segmenting and tagging the original text of the empirical prescription to obtain a sequence of lexical units containing the names of medicinal materials, dosages, diagnostic terms, and operational instructions, with each lexical unit vector corresponding to a unique part-of-speech tag; extracting operational instructions based on a preset TCM behavior template matrix to generate a confidence vector for each lexical unit as a behavioral keyword; aligning the confidence vectors with preset syndrome entity tag results, based on the diagnostic thinking of renowned TCM doctors, to extract semantic information on the scope of application and compatibility principles of syndrome differentiation; mapping special decoction operations to quantitative parameters, where first decoction is mapped to operation level 1, main decoction to operation level 2, and subsequent decoction to operation level 3, and converting the description of heat intensity into quantitative heat intensity levels to form standardized decoction operation parameters. This module uses natural language processing technology specifically for the field of TCM to achieve the structured and parameterized transformation of non-standardized empirical prescription information, ensuring the accurate transmission of core diagnostic and treatment information.

[0016] 3. Multi-dimensional feature evidence storage module

[0017] This method is used to collect standard sample images and physical images of each core medicinal material in the empirical formula. Visual feature vectors are extracted through an image recognition model and matched and verified with the embedded vectors of the standard medicinal materials to generate a consistency verification result and a set of visual feature hashes. At the same time, typical case data and efficacy evaluation indicators corresponding to the empirical formula are collected. After homomorphic encryption processing, a set of clinical feature hashes is extracted to realize the feature collection, authenticity verification and hash extraction of the three-dimensional data of the empirical formula, which includes "text-physical-clinical". The image recognition model consists of four convolutional layers and two fully connected layers. The processing flow of this module includes: acquiring high-resolution images of core medicinal materials using an industrial-grade camera under a fixed light source; binding the images to the medicinal material serial number and the unique ID of the empirical formula; designing an asymmetric channel attention term based on visual main feature representation; obtaining channel attention vectors through channel average pooling followed by a linear transformation; dynamically weighting the response intensities of different channels to construct image recognition vectors; calculating the similarity score between the image recognition vector and the standard medicinal material embedding vector; introducing an image gradient variability regularization term to reduce the risk of false high similarity matching; determining a successful match when the similarity score is greater than a threshold of 0.85; otherwise, triggering a manual review process; serializing the verified image recognition vectors and calculating their hash values ​​to form the visual feature hash set. In the clinical feature evidence storage stage, homomorphic encryption technology is used to protect the privacy of clinical data. The extracted clinical feature hash set, the medicinal material visual feature hash set, and the standardized empirical formula parameter summary values ​​are jointly encapsulated into a structured evidence storage data block, providing multi-dimensional evidence for the reliable storage of empirical formulas.

[0018] 4. Blockchain Hierarchical Storage Network

[0019] Composed of a public blockchain and a permissioned blockchain, this system encapsulates the identity hash credentials of renowned practitioners, standardized empirical formula parameter digest values, herbal visual feature hash sets, and clinical feature hash sets into structured evidence storage data blocks. The permissioned blockchain stores the complete encrypted data of the empirical formula, the original materials of the renowned practitioner's qualifications, and the user access permission list. The public blockchain stores the evidence storage data block hashes, the empirical formula's basic index information, ownership change records, and consensus verification results, achieving the hierarchical management goal of "encrypted storage of privacy data and traceable sharing of public information." This network is configured with smart contracts, whose functions include verifying the integrity of the evidence storage data blocks, auditing access permissions, and recording ownership changes. When an authorized user requests access to the complete empirical formula data, they must first pass the permission verification by the permissioned blockchain node. After successful verification, the data integrity is compared based on the evidence storage data block hashes stored on the public blockchain to ensure that the data has not been tampered with. The permissioned blockchain is deployed at core institutions such as the State Administration of Traditional Chinese Medicine and key TCM hospitals, with only authorized nodes allowed to participate in data reading and writing. The public blockchain adopts a consortium blockchain architecture, incorporating regulatory agencies, industry associations, and authorized medical institutions as nodes to ensure the security and traceability of data storage.

[0020] 5. Intelligent sharing and matching module

[0021] This module is used to analyze the clinical needs of authorized users based on a preset permission management strategy. It achieves intelligent matching of clinical needs with empirical prescriptions through a traditional Chinese medicine (TCM) knowledge graph constructed using a graph convolutional neural network, generating a matching score and personalized application suggestions. It serves as a bridge connecting the evidence of empirical prescriptions with clinical application. The processing flow of this module includes: performing word segmentation and semantic analysis on the clinical need information input by authorized users, such as disease type, syndrome differentiation, and patient constitution characteristics, to extract core need keywords; retrieving empirical prescriptions that match the core need keywords based on the TCM knowledge graph, which includes related information such as TCM attributes, compatibility relationships, syndrome correspondences, and alternative medicinal materials; calculating the matching score between clinical needs and the applicable scope of the empirical prescription, and generating a Top-N candidate empirical prescription list by combining historical clinical application effect statistics of the empirical prescription; and generating personalized application suggestions for each candidate empirical prescription, including dosage adjustment ranges, key points for optimizing decoction operations, alternative medicinal materials, and clinical application precautions. This module only grants search access to authorized users such as qualified medical institutions, licensed physicians, and researchers. It lowers the threshold for clinical application of empirical prescriptions through precise intelligent matching technology, thereby improving application effectiveness and conversion efficiency.

[0022] 6. Clinical Feedback Iteration Module

[0023] This module is used to collect actual implementation data, efficacy data, and optimization suggestions after the clinical application of empirical formulas. It calculates the consistency score and mutation stability index between the actual application and standardized empirical formula parameters, filters valid feedback information, encrypts it, and writes it into the blockchain to form empirical formula iteration update credentials. This is the core of achieving dynamic optimization of empirical formulas. The actual implementation data collected in this module includes the dosage of medicinal materials, decoction temperature curve, changes in heat level, medication duration, and operation timestamps. The processing flow includes: using a soft interval tolerance mechanism to compare the differences between the actual implementation data and standardized empirical formula parameters; introducing a heat level importance function to assign different strictness weights to different heat levels; calculating a consistency score ranging from 0 to 1, with a score closer to 1 indicating better compliance with standards; calculating the temperature change rate between adjacent time points based on continuous time series temperature sampling points; introducing a mutation adjustment coefficient to construct a mutation stability index, with an index close to 1 indicating a stable application process, while a value far less than 1 suggests abnormal operation or medicinal material quality issues; and filtering feedback information with a consistency score higher than the threshold of 0.7 and a mutation stability index close to 1 as valid feedback, generating a set of valid feedback feature hashes. This module works in conjunction with the experience formula structured parsing module. When the number of valid feedback messages reaches a preset threshold, the experience formula structured parsing module optimizes and adjusts the standardized experience formula parameters based on the valid feedback messages. The optimized standardized experience formula parameters are rehashed, associated with the original evidence storage data block, and written into the blockchain to form an experience formula version iteration record, realizing dynamic optimization of the experience formula. At the same time, this module works in conjunction with the multi-dimensional feature evidence storage module. Abnormal feedback messages screened by the clinical feedback iteration module trigger the secondary verification process of the multi-dimensional feature evidence storage module, which re-acquires and matches images of the clinically used medicinal materials and re-verifies them. The verification results are written into the blockchain as the basis for tracing the source of abnormal issues.

[0024] 7. Experienced Party Ownership Protection Module

[0025] This extended functional module of the system, linked with a blockchain-based hierarchical storage network, records the sharing authorization records, usage statistics, and revenue distribution information of the practitioners. Through smart contracts, it automatically executes preset revenue distribution rules and writes the distribution results into the blockchain, creating an immutable distribution certificate. This module enables full recording of changes in the ownership of practitioners, ensuring the legality and traceability of such changes. Simultaneously, through smart contracts, it automatically calculates and distributes the revenue generated by practitioners, fully protecting the intellectual property rights and legitimate interests of renowned traditional Chinese medicine practitioners and providing rights protection for the compliant sharing of knowledge.

[0026] The various modules work together to form a complete technical system, from confirming the qualifications of renowned practitioners, standardizing the analysis of experienced prescriptions and storing evidence in multiple dimensions, to blockchain hierarchical storage, intelligent sharing and matching, and then to clinical feedback iteration and ownership protection. Through clinical feedback, a closed loop of iterative optimization of experienced prescriptions is formed, providing a systematic technical solution for the inheritance, sharing and clinical transformation of the experienced prescriptions of renowned Chinese medicine practitioners.

[0027] Beneficial effects:

[0028] Achieve credible ownership verification and multi-dimensional evidence preservation for experienced practitioners: Complete the ownership verification of renowned practitioners' qualifications through multi-source cross-verification, generate unique identity hash certificates, and solve the problem of ownership ambiguity; Combine the three-dimensional feature preservation of text, physical medicinal materials, and clinical data to achieve comprehensive authenticity verification of experienced practitioners and avoid data tampering and information distortion.

[0029] Achieving standardized analysis and precise inheritance of empirical prescriptions: Through natural language processing technology specific to traditional Chinese medicine, vague empirical prescription information is transformed into standardized and quantifiable parameters, ensuring the accurate transmission of core information such as diagnostic reasoning, compatibility ratios, and decoction procedures, thereby improving the accuracy of inheritance.

[0030] Balancing the openness and privacy of shared expertise: A hierarchical storage architecture of public blockchain and permissioned blockchain is adopted. Privacy data is encrypted and stored on the permissioned blockchain, while public information is traceable on the public blockchain. Combined with a fine-grained permission management strategy, this approach enables extensive sharing of expertise within authorized scope while effectively protecting the core experience and privacy data of renowned traditional Chinese medicine practitioners, breaking down data silos between medical institutions.

[0031] Improving the efficiency and accuracy of clinical translation of empirical formulas: Based on the intelligent matching technology of traditional Chinese medicine knowledge graph, it realizes the precise connection between clinical needs and empirical formulas. Combined with personalized application suggestions, it lowers the threshold for clinical application of empirical formulas and improves application effect and translation efficiency.

[0032] Constructing a dynamic iterative optimization closed loop for empirical formulas: Through the collection, verification, and screening of clinical feedback data, the empirical formulas are dynamically optimized, enabling them to be continuously improved in clinical application, fully leveraging their medical and scientific research value, and promoting the inheritance and innovation of traditional Chinese medicine.

[0033] Achieving full-process traceability and anomaly tracing for experience providers: The immutability of blockchain ensures that the entire process of experience provider operations, from ownership confirmation, evidence storage, sharing to application and iteration, is traceable. The linkage mechanism of each module enables rapid tracing of abnormal issues in clinical applications, providing technical support for the quality control of experience providers. Attached Figure Description

[0034] Figure 1 This is a schematic diagram of the framework of the present invention. Detailed Implementation

[0035] To make the technical solution of the present invention clearer, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0036] Reference Figure 1 As shown, the blockchain-based system for storing and sharing the experience and prescriptions of renowned traditional Chinese medicine practitioners includes:

[0037] The expert qualification verification module is used to collect the identity information, professional qualification certificates, academic lineage and achievement certification materials of famous Chinese medicine practitioners. It completes the qualification confirmation through a multi-source cross-verification mechanism and generates an expert identity hash certificate containing a digital signature, which serves as the core basis for the ownership determination of the expert.

[0038] The structured parsing module for empirical prescriptions is used to obtain the original text, diagnostic reasoning, and key points of decoction techniques of empirical prescriptions from renowned traditional Chinese medicine practitioners. Through word segmentation, terminology alignment, and behavioral instruction extraction, it generates standardized empirical prescription parameters that include the composition of medicinal materials, compatibility ratios, applicable scope of diagnosis, special decoction operations, and contraindications. A hash algorithm is then used to generate a summary value for the standardized empirical prescription parameters.

[0039] The multi-dimensional feature storage module is used to collect standard sample images and physical images of each core medicinal material in the empirical prescription. It extracts visual feature vectors through an image recognition model and matches and verifies them with the embedded vectors of the standard medicinal materials to generate a medicinal material consistency verification result and a set of visual feature hashes. At the same time, it collects typical case data and efficacy evaluation indicators corresponding to the empirical prescription, and extracts a set of clinical feature hashes after homomorphic encryption.

[0040] The blockchain hierarchical storage network consists of a public chain and a permissioned chain. It is used to encapsulate the identity hash certificate of famous experts, the parameter summary value of standardized experience prescriptions, the hash set of visual features of medicinal materials, and the hash set of clinical features into structured evidence storage data blocks. The permissioned chain stores the complete encrypted data of the experience prescription, the original materials of the famous expert's qualifications and the list of user access permissions, while the public chain stores the hash of the evidence storage data block, the basic index information of the experience prescription, the ownership change record and the consensus verification result.

[0041] The intelligent sharing and matching module is used to analyze the clinical needs of authorized users based on preset permission management strategies. It realizes intelligent matching of clinical needs and empirical prescriptions through a traditional Chinese medicine knowledge graph constructed by graph convolutional neural network, and generates matching degree scores and personalized application suggestions.

[0042] The clinical feedback iteration module is used to collect actual implementation data, efficacy data and optimization suggestions after the clinical application of empirical formulas, calculate the consistency score and mutation stability index between actual application and standardized empirical formula parameters, screen valid feedback information and write it into the blockchain after encryption processing to form empirical formula iteration update vouchers.

[0043] Example 1

[0044] Reference Figure 1 As shown, the blockchain-based system for storing and sharing the experience prescriptions of renowned traditional Chinese medicine practitioners, taking a prescription for eczema from a famous practitioner as an example, operates as follows:

[0045] Step 1: Verification and Confirmation of Qualifications of Renowned Experts

[0046] National TCM masters submit qualification documents such as their ID card, practicing physician certificate, national master certification, academic lineage, and clinical achievement certification materials through the system. The master qualification verification module connects to the National Administration of Traditional Chinese Medicine's practicing physician database and the TCM Association certification system to complete the authoritative verification of qualification data. The master's handwritten signature and personal seal are digitally signed and stored to generate a signature hash value. The master's qualifications are verified by distributed signatures from master-disciple nodes in the lineage, forming a multi-node consensus. The module integrates the qualification verification data hash, signature hash value, and lineage node verification signature to generate a unique master identity hash certificate, completing the confirmation of master qualification. The certificate is written to the permissioned chain of the blockchain hierarchical storage network.

[0047] Step 2: Structured Analysis of Empirical Methods

[0048] The original text of a renowned expert's eczema prescription, along with explanations of the diagnostic approach (damp-heat accumulation type eczema) and key points of the decoction technique, were submitted. The prescription's structured parsing module segmented and annotated the original text, extracting the names and dosages of herbs such as Astragalus membranaceus, Sophora flavescens, and Kochia scoparia, as well as operational instructions such as "decoct first" and "simmer over low heat for 20 minutes." Based on a TCM behavioral template matrix, a confidence vector of operational keywords was generated, and semantic alignment was achieved by combining it with the "damp-heat accumulation type" syndrome label. "Decoct first" was mapped to operation level 1, and "simmer over low heat" was converted into a quantified heat level, generating standardized prescription parameters that include the herbal composition, compatibility ratio, applicable scope (damp-heat accumulation type eczema), standardized decoction operation, and contraindications (avoid spicy and cold foods). The SHA-256 algorithm was used to generate a summary value for the standardized prescription parameters.

[0049] Step 3: Multidimensional Feature Storage

[0050] The multi-dimensional feature storage module acquires high-resolution images of core medicinal materials such as Astragalus membranaceus, Sophora flavescens, and Kochia scoparia using industrial-grade cameras. Each image is bound to a serial number and a unique ID for the empirical formula. The medicinal material images are input into an image recognition model to extract the main visual features and construct an image recognition vector by combining it with asymmetric channel attention terms. This vector is then matched against the standard medicinal material embedding vectors for similarity. A similarity score higher than 0.85 indicates a successful match. The image recognition vectors are serialized and hash values ​​are calculated to generate a set of visual feature hashes for the medicinal materials. Data from 10 typical eczema cases corresponding to the empirical formula, along with efficacy evaluation indicators, are collected. After homomorphic encryption, clinical features are extracted to generate a set of clinical feature hashes. The module synchronizes the medicinal material consistency verification results, the visual feature hash set, and the clinical feature hash set to a blockchain hierarchical storage network.

[0051] Step 4: Blockchain-based hierarchical evidence storage

[0052] The blockchain hierarchical storage network encapsulates the expert's identity hash certificate, standardized empirical prescription parameter summary value, medicinal material visual feature hash set, and clinical feature hash set into a structured evidence storage data block; stores the complete encrypted data of the empirical prescription, the original materials of the expert's qualifications, and the initial access permission list on the permissioned chain; writes the evidence storage data block hash, the empirical prescription basic index (eczema, damp-heat accumulation type), and ownership information into the public chain; the smart contract completes the integrity verification of the evidence storage data block, generates a successful evidence storage certificate, and feeds it back to the expert and the system management end.

[0053] Step 5: Smart Shared Matching

[0054] A dermatologist at a top-tier traditional Chinese medicine hospital submitted an access request through the system due to clinical needs, specifying the application scenario as "clinical diagnosis and treatment of damp-heat eczema." After verifying the physician's and medical institution's authorization qualifications, the system granted access to retrieve prescriptions. The physician entered the clinical need "damp-heat eczema, patient with a weak constitution" into the intelligent sharing and matching module. The module performed semantic analysis on the need, extracting core keywords. Based on the knowledge graph of traditional Chinese medicine, it retrieved and matched prescriptions. This famous eczema prescription had a matching score of 92 points, ranking first in the candidate list. The module generated personalized application suggestions, including "increase the dosage of Astragalus membranaceus by 10g (suitable for a weak constitution)," "extend the simmering time by 5 minutes," and "avoid taking it with seafood," and pushed these suggestions to the physician.

[0055] Step 6: Clinical Application and Feedback Data Collection

[0056] Physicians refer to personalized application suggestions and use this empirical formula in clinical practice. The system collects actual execution data of clinical application through edge computing nodes: dosage of Astragalus membranaceus after adjustment, decoction temperature curve, changes in heat level, medication duration, operation timestamp, etc. At the same time, it collects the patient's efficacy data (eczema reduction rate of 85%) and records of no adverse reactions. Physicians submit optimization suggestions such as "the dosage of Sophora flavescens can be appropriately reduced for pediatric patients to reduce irritation".

[0057] Step 7: Clinical Feedback Validation and Iteration

[0058] The clinical feedback iteration module processes the collected actual execution data, efficacy data, and optimization suggestions; it calculates a consistency score of 0.88 using a soft interval tolerance mechanism, which is higher than the threshold of 0.7; based on temperature sampling points, the mutation stability index is calculated to be close to 1, indicating that the application process is stable; the feedback information is selected as valid feedback, and after encryption, a set of valid feedback feature hashes is generated; when the number of valid feedback information for the empirical formula reaches 20 preset thresholds, the clinical feedback iteration module, in conjunction with the empirical formula structured parsing module, optimizes the standardized empirical formula parameters based on the valid feedback, adding the content of "adjustment range of Sophora flavescens dosage for pediatric patients"; the optimized standardized empirical formula parameters regenerate summary values, are associated with the original evidence data block and written to the blockchain, forming an empirical formula V2.0 version iteration record, realizing dynamic optimization of the empirical formula.

[0059] Step 8: Ownership Protection and Profit Distribution

[0060] The experience provider ownership protection module records the experience provider's shared authorization record (the tertiary TCM hospital is the authorized user) and the number of clinical uses (cumulative 50 times); through the revenue distribution rules preset by the smart contract, it automatically calculates the TCM hospital's usage revenue and completes the distribution, and the distribution result is written into the blockchain to form an immutable distribution certificate; at the same time, the module records the experience provider's version iteration, ownership changes and other information to realize full-process rights management.

[0061] The embodiments described above are merely illustrative of several implementations of the present invention, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of the present invention. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the present invention, and these modifications and improvements all fall within the scope of protection of the present invention. Therefore, the scope of protection of this patent should be determined by the appended claims.

Claims

1. A blockchain-based system for storing and sharing the experience prescriptions of renowned traditional Chinese medicine practitioners, comprising a module for verifying the qualifications of practitioners, a module for structured parsing of experience prescriptions, a module for storing multi-dimensional features, a blockchain hierarchical storage network, an intelligent sharing and matching module, and a clinical feedback and iteration module, characterized by: The aforementioned expert qualification verification module is used to collect the identity information, professional qualification certificates, academic lineage and achievement certification materials of renowned Chinese medicine practitioners. It completes the qualification confirmation through a multi-source cross-verification mechanism and generates an expert identity hash certificate containing a digital signature, which serves as the core basis for the ownership determination of the expert. The structured parsing module for the empirical formula is used to obtain the original text, diagnostic reasoning, and key points of decoction techniques of the empirical formulas of famous Chinese medicine practitioners. Through word segmentation, terminology alignment, and behavior instruction extraction, it generates standardized empirical formula parameters that include the composition of medicinal materials, compatibility ratio, scope of application for diagnosis, special decoction operations, and contraindications. A hash algorithm is then used to generate a summary value of the standardized empirical formula parameters. The multi-dimensional feature storage module is used to collect standard sample images and physical images of each core medicinal material in the empirical prescription, extract visual feature vectors through an image recognition model and match and verify them with the standard medicinal material embedding vectors to generate a medicinal material consistency verification result and a visual feature hash set; at the same time, it collects typical case data and efficacy evaluation indicators corresponding to the empirical prescription, and extracts a clinical feature hash set after homomorphic encryption processing. The blockchain hierarchical storage network consists of a public chain and a permissioned chain. It is used to encapsulate the expert identity hash certificate, standardized empirical formula parameter summary value, medicinal material visual feature hash set, and clinical feature hash set into structured evidence storage data blocks. The permissioned chain stores the complete encrypted data of the empirical formula, the original materials of the expert's qualifications and the user access permission list. The public chain stores the evidence storage data block hash, the empirical formula basic index information, ownership change records and consensus verification results. The intelligent sharing and matching module is used to analyze the clinical needs of authorized users based on a preset permission management strategy, and realize the intelligent matching of clinical needs and empirical prescriptions through a traditional Chinese medicine knowledge graph constructed by a graph convolutional neural network, generating a matching score and personalized application suggestions. The clinical feedback iteration module is used to collect actual implementation data, efficacy data and optimization suggestions after the clinical application of the empirical formula, calculate the consistency score and mutation stability index between the actual application and the standardized empirical formula parameters, screen the effective feedback information and write it into the blockchain after encryption, forming an empirical formula iteration update certificate.

2. The blockchain-based system for storing and sharing the experience prescriptions of renowned traditional Chinese medicine practitioners as described in claim 1, characterized in that: The multi-source cross-validation mechanism of the expert qualification verification module includes: Connect with the National Administration of Traditional Chinese Medicine's database of licensed physicians and the certification system for academic organizations in traditional Chinese medicine to obtain authoritative qualification verification data; Digital signature technology is used to digitally preserve and generate signature hash values ​​for handwritten signatures and seal marks of famous figures. Distributed verification is carried out through the master-apprentice relationship chain nodes in the academic lineage, and consensus is formed by each lineage node signing and confirming the qualification information. The identity hash certificate of the famous person is generated by combining the qualification verification data hash, the signature hash value, and the inheritance node verification signature.

3. The blockchain-based system for storing and sharing the experience prescriptions of renowned traditional Chinese medicine practitioners as described in claim 1, characterized in that, The processing flow of the empirical square structured parsing module includes: The original text of the empirical prescription is processed by word segmentation and term tagging to obtain a word sequence containing the name of medicinal material, dosage, syndrome differentiation terminology and operation instructions. Each word vector corresponds to a unique part-of-speech tag. Operation instructions are extracted based on a preset TCM behavior template matrix, and each word element is generated as a confidence vector of the behavior keyword. Combining the diagnostic reasoning of renowned traditional Chinese medicine practitioners, the confidence vector is aligned with the preset syndrome entity label results to extract semantic information on the applicable scope and compatibility principles of syndrome differentiation. Special decoction operations are mapped to quantitative parameters, with the first decoction mapped to operation level 1, the main decoction mapped to operation level 2, and the subsequent decoction mapped to operation level 3. The description of heat is converted into a quantitative heat level, forming standardized decoction operation parameters.

4. The blockchain-based system for storing and sharing the experience prescriptions of renowned traditional Chinese medicine practitioners as described in claim 1, characterized in that: In the multi-dimensional feature storage module, the image recognition model consists of four convolutional layers and two fully connected layers. The processing flow of this module includes: High-resolution images of core medicinal materials are acquired using an industrial-grade camera under a fixed light source, and the images are bound to the serial number of the medicinal material and the unique ID of the empirical formula. Based on visual main feature representation, an asymmetric channel attention term is designed. After channel average pooling, a linear transformation is performed to obtain the channel attention vector. The response intensity of different channels is dynamically weighted to construct the image recognition vector. The similarity score between the image recognition vector and the standard medicinal material embedding vector is calculated. An image gradient variability regularization term is introduced to reduce the risk of false high similarity matching. When the similarity score is greater than the threshold of 0.85, the match is considered successful; otherwise, a manual review process is triggered. The image recognition vectors that pass the verification are serialized and their hash values ​​are calculated to form the visual feature hash set.

5. The blockchain-based system for storing and sharing the experience prescriptions of renowned traditional Chinese medicine practitioners as described in claim 1, characterized in that, The blockchain hierarchical storage network is configured with smart contracts. The functions of the smart contracts include verifying the integrity of the evidence storage data blocks, auditing access permissions, and recording ownership changes. When an authorized user applies to access the complete data of the knowledge party, the user must first pass the permission verification of the permission chain node. After the verification is passed, the data integrity is compared based on the hash of the evidence storage data block stored on the public chain to ensure that the data has not been tampered with.

6. The blockchain-based system for storing and sharing the experience prescriptions of renowned traditional Chinese medicine practitioners as described in claim 1, characterized in that, The processing flow of the intelligent sharing matching module includes: The system performs word segmentation and semantic analysis on clinical demand information such as disease type, syndrome classification, and patient physical characteristics input by authorized users to extract core demand keywords. Based on the experience formula of traditional Chinese medicine knowledge graph retrieval and core demand keyword matching, the traditional Chinese medicine knowledge graph contains related information such as the attributes of traditional Chinese medicine, compatibility relationship, syndrome correspondence relationship, and alternative solutions of medicinal materials; Calculate the matching score between clinical needs and the applicable scope of empirical formulas, and generate a Top-N candidate empirical formula list by combining the historical clinical application effect statistics of empirical formulas. Personalized application suggestions are generated for each candidate empirical formula, including dosage adjustment range, key points for optimizing decoction operation, alternative medicinal materials, and precautions for clinical application.

7. The blockchain-based system for storing and sharing the experience prescriptions of renowned traditional Chinese medicine practitioners as described in claim 1, characterized in that, The actual execution data collected by the clinical feedback iteration module includes the dosage of medicinal materials, decoction temperature curve, changes in heat level, medication duration, and operation timestamp. The processing flow of this module includes: A soft interval tolerance mechanism is used to compare the differences between actual execution data and standardized empirical parameters. A temperature importance function is introduced to assign different strictness weights to different temperature levels. A consistency score with a value range of 0-1 is calculated. The closer the score is to 1, the more the actual application conforms to the standard. The rate of temperature change between adjacent time points is calculated based on temperature sampling points of continuous time series. A mutation regulation coefficient is introduced to construct a mutation stability index. An index close to 1 indicates that the application process is stable, while an index much less than 1 indicates abnormal operation or quality problems of medicinal materials. Feedback information with a consistency score higher than the threshold of 0.7 and a mutation stability index close to 1 is selected as valid feedback, and a set of valid feedback feature hashes is generated.

8. The blockchain-based system for storing and sharing the experience prescriptions of renowned traditional Chinese medicine practitioners as described in claim 1, characterized in that, It also includes an experience party ownership protection module, which is linked to the blockchain hierarchical storage network. Its function is to record the experience party's shared authorization records, usage statistics and revenue distribution information. Through smart contracts, it automatically executes preset revenue distribution rules and writes the revenue distribution results into the blockchain to form an immutable distribution certificate.

9. The blockchain-based system for storing and sharing the experience prescriptions of renowned traditional Chinese medicine practitioners according to claim 1, characterized in that, The clinical feedback iteration module works in conjunction with the empirical formula structured parsing module. When the effective feedback information reaches a preset threshold, the empirical formula structured parsing module optimizes and adjusts the standardized empirical formula parameters based on the effective feedback information. After the optimized standardized empirical formula parameters are rehashed, they are associated with the original evidence storage data block and written into the blockchain to form an empirical formula version iteration record, thereby realizing the dynamic optimization of the empirical formula.

10. The blockchain-based system for storing and sharing the experience prescriptions of renowned traditional Chinese medicine practitioners according to claim 1, characterized in that, The multi-dimensional feature storage module works in conjunction with the clinical feedback iteration module. Abnormal feedback information filtered out by the clinical feedback iteration module triggers a secondary verification process in the multi-dimensional feature storage module, which re-acquires and matches images of the actual medicinal materials used in clinical applications. The verification results are written into the blockchain as a basis for tracing the source of abnormal issues.

Images