Orthopedic implant patent value evaluation system

Abstract

The technical scheme of the invention discloses an orthopedic implant patent value evaluation system, which is characterized by comprising an orthopedic implant evaluation reference standard data acquisition module, an orthopedic implant patent data integration module, a multi-source data feature construction module, an orthopedic implant patent comprehensive value evaluation model and an evaluation and result output module. The invention provides an orthopedic implant series patent value evaluation and analysis system for identifying high-quality, convertible and investable orthopedic implant patents, and can combine legal, economic, technical and other patent evaluation indexes with clinical research data, supervision systems, clinical guidelines and the like. Dynamic evaluation and prospective prediction are carried out on the value of a series of orthopaedic implant patents, market data and a machine learning algorithm are introduced, an industrial exclusive evaluation platform is constructed, a system value of achievement transformation-patent value evaluation is established, and scientificity, adaptability and practicability of an evaluation result are ensured.

Description

Technical Field

[0001] This invention relates to a system for evaluating the patent value of orthopedic implant patents. Background Technology

[0002] Bone is a type of connective tissue with a unique structure in the human body. Based on its structure and distribution, it can be divided into two main categories: cancellous bone and cortical bone. Due to its strong regenerative capacity, especially in children and young adults, minor bone cracks and fractures can heal spontaneously within a short time without much external intervention. However, for patients with significant bone defects, such as those who have undergone bone tumor removal or have poor bone tissue healing ability, it is necessary to implant repair materials to guide bone tissue regeneration. These implanted materials are called orthopedic implants. Orthopedic implants include metal plates, screws, intramedullary nails, artificial joints, and other non-metallic materials. As a high-risk Class III medical device, they are widely used clinically in the treatment of diseases.

[0003] The number of patents for orthopedic implants has surged in recent years, but the quality of these patents varies greatly, and their commercialization outcomes differ. Current patent valuation index systems primarily extract evaluation indicators from the legal, technical, and economic dimensions of the patent itself, employing qualitative and quantitative tools for evaluation and analysis. Commonly used analytical methods include asset valuation theories such as the cost approach, market approach, and income approach. These methods treat patents as intangible assets, quantifying their value in monetary terms. Specifically, the cost approach focuses on the replacement cost of the patent-related technology; the market approach relies on comparable market transactions of the patented technology or product; and the income approach calculates the future expected revenue of the patented technology or product by discounting it. Current research on patent valuation index systems exhibits different focuses among various stakeholders: research institutions (e.g., universities, government departments, hospitals) emphasize the index evaluation system itself; asset appraisal companies focus on asset returns; and patent agencies focus on the entire patent lifecycle. This research suffers from problems such as strong subjectivity in evaluation indicators, difficulty in quantification, and insufficient industry universality and adaptability. Furthermore, in specialized fields, limited data resources hinder the further improvement and development of patent valuation index systems.

[0004] In the patent evaluation process for orthopedic implants, it is crucial to consider both clinical and regulatory factors. Patent value depends not only on technical features but also on factors throughout the entire lifecycle, including clinical trials, registration approval, and post-market surveillance. Furthermore, key indicators such as the biocompatibility, mechanical compatibility, stress shielding effect, and long-term failure impact of the implant material must be emphasized. Finally, the impact of national centralized procurement, hospital bidding, medical insurance price lists, and price negotiations on commercialization prospects must be considered, along with factors such as surgical techniques, usage habits, and patient preferences. However, the current patent value evaluation system cannot meet all these requirements. Summary of the Invention

[0005] The present invention aims to solve the following technical problems:

[0006] (1) The existing patent evaluation index system has weak data capture capabilities and cannot match and analyze patent data with clinical guidelines, clinical trials, registration and approval systems, etc.

[0007] (2) The existing patent evaluation indicators are not broad enough, and there is a lack of quantitative evaluation indicators for the biological and mechanical properties of orthopedic implants.

[0008] (3) The existing patent evaluation index system and evaluation methods are not advanced enough. They cannot integrate multi-source data fusion and machine learning methods, resulting in slow processing efficiency and low accuracy of results.

[0009] To address the aforementioned technical problems, the present invention discloses a patent value assessment system for orthopedic implants, characterized by comprising:

[0010] The orthopedic implant evaluation reference standard data acquisition module is used to acquire and structure key data indicators related to the patent to be evaluated from multi-source heterogeneous data. The multi-source heterogeneous data comes from orthopedic implant-related literature, clinical guidelines, and expert consensus.

[0011] The orthopedic implant patent data integration module is used to efficiently and accurately collect patent-related data of the patents to be evaluated from multiple heterogeneous data sources, and to perform in-depth cleaning, sorting and feature extraction.

[0012] The multi-source data feature construction module is used to construct multi-source data features of the patent to be evaluated based on the key data indicators obtained by the orthopedic implant evaluation reference standard data acquisition module and the patent-related data obtained by the orthopedic implant patent data integration module, forming a comprehensive feature vector that is finally used as input for the subsequent orthopedic implant patent comprehensive value assessment model.

[0013] The comprehensive value assessment model for orthopedic implant patents takes the multi-source data features D obtained by the multi-source data feature construction module as input, and uses five sub-models to calculate five dimensions: clinical and scientific research, policy and regulation, patent legal value, patent technical value and patent economic value. The evaluation data of each dimension is calculated, and then the outputs of the five sub-models are fused through a gating network model.

[0014] The evaluation and results output module is used to analyze, interpret, and visualize the results output by the comprehensive value assessment model for orthopedic implant patents, and finally generate an intuitive patent value assessment report for all patents to be evaluated.

[0015] Preferably, the key data indicators include family size / geographical coverage data, legal status / litigation risk data, claim coverage, patent citation network analysis data, core patent expiration date, international standard adoption data, industry technical specification compliance data, improvement in major endpoints, 10-year revision rate, functional score change value, adverse event incidence rate, complication rate reduction, international guideline recommendation strength data, indication coverage consistency data, CE / FDA / NMPA certification validity data, label indication compliance data, inclusion in the national medical insurance catalog data, reimbursement ratio / out-of-pocket amount, historical winning bid price trend data, procurement volume market share data, target patient penetration rate, unit price × usage prediction model, BOM cost structure breakdown data, industry average gross profit comparison data, tertiary hospital coverage rate, channel cooperation stability data, satisfaction score, pain relief VAS score, ease of operation score, and intraoperative adaptation problem incidence rate.

[0016] Preferably, the orthopedic implant evaluation reference standard data acquisition module systematically sorts out the collected data and extracts text through regular expressions and rule matching, or uses Docano for manual annotation combined with natural language processing tools such as SpaCy for entity recognition and relation extraction, transforming unstructured policy text into structured data.

[0017] Preferably, the orthopedic implant patent data integration module includes an orthopedic implant patent multi-source heterogeneous data acquisition unit and an orthopedic implant patent data cleaning and sorting unit, wherein:

[0018] The orthopedic implant patent multi-source heterogeneous data acquisition unit is used to acquire multi-source heterogeneous data indicators of all orthopedic implant patents related to the patent to be evaluated.

[0019] The orthopedic implant patent data cleaning and sorting unit is used to perform in-depth cleaning, format conversion and structural integration of the original patent data obtained by the orthopedic implant patent multi-heterogeneous data acquisition unit.

[0020] Preferably, the patent data indicators collected by the orthopedic implant patent multivariate heterogeneous data acquisition unit include applicant, inventor, patent type, patent field, full text of technical specification, publication number, priority, patent family information, legal status, citation relationship, assignment / license record, litigation information, communication standard, designated country patent number, and expected expiration date.

[0021] Preferably, the data cleaning process adopted by the orthopedic implant patent data cleaning and sorting unit specifically includes the following steps:

[0022] Step 1, Data Retrieval and Preliminary Storage: The collected original patent data is initially stored according to fields;

[0023] Step 2, Data Exploration: Perform descriptive statistical analysis on the data to identify missing values, outliers, duplicates, and inconsistent formats;

[0024] Step 3, Field Cleaning: Deduplicat key fields, standardize their format, and correct any errors;

[0025] Step 4, Code and Category Mapping: Map the category codes in the original database to a unified category system;

[0026] Step 5, Ontology and Lexical Standardization: Standardize professional terms and normalize synonyms for text fields such as patent specifications and abstracts;

[0027] Step 6: Structured text extraction: Extract semi-structured or structured information from the patent text using rule matching, regular expressions, or natural language processing techniques.

[0028] Preferably, the multi-source data feature is represented as D, then:

[0029]

[0030] in, For feature data that can represent the value of patented technology, For characteristic data that can represent the economic value of a patent, For feature data that can represent the legal value of a patent, To be able to represent the characteristic data of clinical and scientific research To be able to represent the characteristic data of policies and regulations.

[0031] Preferably, the comprehensive value assessment model for orthopedic implant patents is represented by E. β (·), then we have:

[0032]

[0033] Among them, G β (·) represents the computational model, D tech (·) represents a sub-model used to obtain data for evaluating the value dimension of patented technologies. eco (·) represents a sub-model used to obtain data for evaluating the economic value dimension of patents, D legal (·) represents a sub-model used to obtain data for evaluating the legal value dimension of patents. c&r (·) represents the sub-model used to obtain evaluation data for clinical and research dimensions, D p &s (·) is a sub-model used to obtain evaluation data for policy and regulatory dimensions.

[0034] Preferably, sub-model D tech(·) Based on the rules set in the table below, the final patent technology value dimension evaluation data are obtained from the aspects of core innovation and patent technology depth, technology maturity, safety and effectiveness evidence, long-term stability / durability verification, multi-indication coverage depth, differentiation from existing technologies, and production quality control and scalability.

[0035]

[0036]

[0037]

[0038] Based on the table above, sub-model D tech (·) Obtain the final evaluation score D for the patent technology value dimension. tech The calculation process used is as follows:

[0039]

[0040] Where: the score for the secondary indicator is represented by S. i This indicates that i = 1, 2, 3, 4, ..., 7, which correspond to the scores of the seven secondary indicators in the table above: S1: Core Innovation and Patent Technology Depth Score, S2: Technology Maturity Score, S3: Safety and Efficacy Evidence Score, S4: Long-Term Stability / Durability Verification Score, S5: Multi-Indication Coverage Depth Score, S6: Differentiation from Existing Technologies S7: Production Quality Control and Scalability Score; W i C is the weight of the score of the i-th secondary indicator; cor For correction factors;

[0041] Sub-model D eco (·) Based on the rules set in the table below, the final evaluation data of the patent economic value dimension is obtained from the aspects of potential market size, medical insurance / payment adaptability, centralized procurement adaptability, cost and price competitiveness, commercialization channel maturity, investment / financing attractiveness, and international market expansion potential.

[0042]

[0043]

[0044]

[0045] Based on the table above, sub-model D eco (·) Obtain the final patent economic value dimension score D eco The calculation process used is as follows:

[0046]

[0047] Where: the score for the secondary indicator is represented by S. i This means that i = 1, 2, 3, 4, ..., 6, which correspond to the scores of the 6 secondary indicators in Table 3 above: S1: Medical insurance / payment suitability score, S2: centralized procurement suitability score, S3: cost and price competitiveness score, S4: commercialization channel maturity score, S5: investment / financing attractiveness score, and S6: international market expansion potential score.

[0048] Sub-model D legal (·) Based on the rules set in the table below, the final evaluation data of the patent legal value dimensions are obtained from the aspects of patent family breadth, remaining protection period, claim coverage, patent legal status, citation influence, procedural security, freedom of implementation, and transferability / licensability potential;

[0049]

[0050]

[0051]

[0052] Based on the table above, sub-model D legal (·) Obtain the final patent legal value dimension score D legal The calculation process used is as follows:

[0053]

[0054] Where: the score for the secondary indicator is represented by S. i The values ​​i = 1, 2, 3, 4, ..., 8 correspond to the scores of the eight secondary indicators in Table 4 above: S1: Patent family breadth score, S2: Remaining protection years score, S3: Claim coverage score, S4: Patent legal status score, S5: Citation influence score, S6: Procedural security score, S7: Freedom of implementation score, and S8: Transferability / licensability potential score.

[0055] Sub-model D c&r (·) Based on the rules set in the table below, the final clinical and research dimension evaluation data were obtained from the following aspects: number of international guideline recommendations, number of domestic guideline recommendations, recommendation level, standard adoption rate, number of academic conference citations, academic visibility, mention rate in comments / meta-analysis, number of patent citations, indication matching degree, improvement in revision rate, achievement rate of prosthesis service life, decrease in complication rate, and improvement in PROMs.

[0056]

[0057]

[0058]

[0059]

[0060] Based on the table above, sub-model D c&r (·) Obtain the final clinical and research dimension score D c&r The calculation process used is as follows:

[0061]

[0062] Where: the score for the secondary indicator is represented by S. i Let i = 1, 2, 3, 4, ..., 13, corresponding to the scores of the 13 secondary indicators in the table above: S1: International guideline recommendation count score, S2: Domestic guideline recommendation count score, S3: Recommendation level score, S4: Standard adoption score, S5: Academic conference citation count score, S6: Academic visibility score, S7: Comment / Meta-analysis mention rate score, S8: Patent citation count score, S9: Indication matching score, S... 10 : Improvement score for renovation rate, S 11 : Prosthetic lifespan achievement rate score, S 12 : Complication rate decrease score, S 13 : Improvement score of PROMs;

[0063] Sub-model D p&s (·) Based on the rules set in the table below, the final policy and regulatory dimension evaluation data will be obtained from aspects such as patient satisfaction, physician operation feedback, quality management system and business compliance, DRG / DIP grouping adaptation, post-marketing policy supervision, ethical review, adverse events and disputes, and innovation / priority review and approval and supporting measures.

[0064]

[0065]

[0066] Based on the table above, sub-model D p&s (·) Obtain the final policy and regulatory dimension score D p&s The calculation process used is as follows:

[0067]

[0068] Where: the score for the secondary indicator is represented by S. iThe values ​​i = 1, 2, 3, 4, ..., 8 correspond to the scores of the eight secondary indicators in the table above: S1: Patient satisfaction score, S2: Physician operation feedback score, S3: Quality management system and business compliance score, S4: DRG / DIP grouping adaptation score, S5: Post-marketing policy supervision score, S6: Ethical review score, S7: Adverse events and disputes score, and S8: Innovation / priority review and approval and supporting score.

[0069] Preferably, the evaluation and result output module includes a value ranking and grade division unit, a result visualization unit, and a patent value analysis report generation unit, wherein:

[0070] The value ranking and grade division unit is used to rank the comprehensive value scores of all patents to be evaluated output by the comprehensive value assessment model for orthopedic implants in percentile order, and calculate the percentile ranking of each sample in descending order.

[0071] The results visualization unit generates multi-dimensional patent value analysis charts based on the evaluation results output by the value ranking and grade division unit and / or the comprehensive value evaluation model for orthopedic implants.

[0072] The patent value analysis report generation unit generates a systematic orthopedic implant patent value analysis report based on the evaluation results output by the value ranking and grading unit and / or the comprehensive value assessment model for orthopedic implant patents.

[0073] This invention proposes a patent value assessment and analysis system for orthopedic implants to identify high-quality, commercially viable, and investable patents. This system integrates legal, economic, and technical patent evaluation indicators with clinical research data, regulatory systems, and clinical guidelines to dynamically assess and predict the value of orthopedic implant patents. It incorporates market data and machine learning algorithms to build an industry-specific assessment platform, establishing a systematic value assessment system for technology transfer and patent value evaluation, ensuring the scientific rigor, adaptability, and practicality of the assessment results. The disclosed value assessment system provides information on the patent value of orthopedic implants, showcasing their value composition from multiple perspectives. It offers a systematic evaluation method for orthopedic implant patents, facilitating comparison and identification of high-value patents by pharmaceutical investors for commercialization investment. Furthermore, this system provides insights for entrepreneurs seeking to commercialize orthopedic implant patents, revealing what constitutes a high-value patent, the current status and value of existing patents, and promoting the invention and selection of more high-value orthopedic implant patents, thus driving the commercialization of orthopedic implant technology. Attached Figure Description

[0074] Figure 1The working process of the system disclosed in this invention is illustrated. Detailed Implementation

[0075] The present invention will be further illustrated below with reference to specific embodiments. It should be understood that these embodiments are for illustrative purposes only and are not intended to limit the scope of the invention. Furthermore, it should be understood that after reading the teachings of this invention, those skilled in the art can make various alterations or modifications to the invention, and these equivalent forms also fall within the scope defined by the appended claims.

[0076] like Figure 1 As shown in the figure, the orthopedic implant patent value assessment system disclosed in this embodiment of the invention includes an orthopedic implant evaluation reference standard data acquisition module, an orthopedic implant patent data integration module, a multi-source data feature construction module, an orthopedic implant patent comprehensive value assessment model, and an evaluation and result output module.

[0077] The Orthopedic Implant Evaluation Reference Standard Data Acquisition Module is used to acquire and structure key data indicators from multi-source heterogeneous data related to orthopedic implants, including literature, clinical guidelines, and expert consensus, to reflect the guiding significance and application potential of the patent technology in clinical practice.

[0078] In one preferred embodiment of the present invention, the key data indicators acquired by the orthopedic implant evaluation reference standard data acquisition module are shown in Table 1 below.

[0079] Table 1. Reference Standards for the Evaluation of Orthopedic Implants

[0080]

[0081] As shown in Table 1 above, the data sources include literature and academics, guidelines and standards, registration and real-world data, regulation and labeling, patents and legal matters, economics and channels, and other aspects. Among them, literature and academics include WOS, PubMed / PMC, and conference proceedings; guidelines and standards include AAOS / NICE / WHO / Chinese Medical Association, ISO / ASTM / YY / group standards; registration and real-world data include joint registrations such as NJR and AOANJRR, hospital HIS / claims (compliance and anonymization); regulation and labeling data include FDA (510(k) / DeNovo / PMA, MAUDE), EUDAMED; patents and legal matters include WIPO, INPADOC, USPTO / EPO / CNIPA, and Darts-IP; economics and channels include centralized procurement portals, medical insurance catalogs, bidding announcements, financial annual reports / prospectuses; and other aspects include relevant information obtained from official authoritative websites such as the website of the National Medical Products Administration, the China Medical Device Information Network, and the website of the National Healthcare Security Administration.

[0082] The data acquisition module for the orthopedic implant evaluation reference standard systematically organizes the collected data and extracts text through regular expressions and rule matching, or uses Docano for manual annotation combined with natural language processing tools such as SpaCy for entity recognition and relation extraction, transforming unstructured policy text into structured data.

[0083] The orthopedic implant patent data integration module is used to efficiently and accurately collect patent-related data of the patents to be evaluated from multiple heterogeneous data sources, and to perform in-depth cleaning, sorting, and feature extraction to provide high-quality input for the upper-level evaluation model. In this embodiment of the invention, the orthopedic implant patent data integration module further includes an orthopedic implant patent multi-source heterogeneous data acquisition unit and an orthopedic implant patent data cleaning and sorting unit.

[0084] The orthopedic implant patent multi-source heterogeneous data acquisition unit is used to acquire multi-source heterogeneous data indicators of all orthopedic implant patents related to the patent to be evaluated. In one preferred embodiment of the invention, the orthopedic implant patent multi-source heterogeneous data acquisition unit primarily uses API interfaces to batch filter and collect series of orthopedic implant-related patent data from professional patent data platforms such as the Incopat database and the PatSnap database. In another preferred embodiment of the invention, the orthopedic implant patent multi-source heterogeneous data acquisition unit uses the patent number as a search condition and supplements it with information from open databases such as Google Patents to ensure data comprehensiveness.

[0085] The patent data indicators collected by the orthopedic implant patent multi-dimensional heterogeneous data acquisition unit include, but are not limited to: applicant, inventor, patent type, patent field (such as IPC classification number), full text of technical specification, publication number, priority, patent family information, legal status, citation relationship (cited / referenced), assignment / licensing records, litigation information, communication standards, designated country patent number, expected expiration date, and other fields.

[0086] The orthopedic implant patent data cleaning and sorting unit is used to perform in-depth cleaning, format conversion and structural integration of the original patent data obtained by the orthopedic implant patent multi-dimensional heterogeneous data acquisition unit, to ensure the high quality and usability of the data, and to provide standardized input for subsequent feature extraction.

[0087] In a preferred embodiment of the present invention, the data cleaning process employed by the orthopedic implant patent data cleaning and sorting unit specifically includes the following steps:

[0088] Step 1, Data Retrieval and Preliminary Storage: The collected original patent data is initially stored according to fields;

[0089] Step 2, Data Exploration: Perform descriptive statistical analysis on the data to identify missing values, outliers, duplicates, and inconsistent formats;

[0090] Step 3, Field Cleaning: Deduplicat key fields (e.g., duplicate patent records), standardize formats (e.g., standardize date formats and unify text encoding), and correct error values;

[0091] Step 4, Code and Classification Mapping: Map the classification codes (such as IPC classifications) in the original database to a unified classification system;

[0092] Step 5, Ontology and Lexical Standardization: Standardize professional terms and normalize synonyms for text fields such as patent specifications and abstracts, for example, "bone nail" and "bone screw".

[0093] Step 6: Structured text extraction: Extract semi-structured or structured information (such as technical solutions, main effects, etc.) from the patent text using rule matching, regular expressions, or natural language processing (NLP) techniques (such as keyword extraction and dependency parsing).

[0094] The multi-source data feature construction module is used to construct multi-source data features of the patent to be evaluated based on the key data indicators obtained by the orthopedic implant evaluation reference standard data acquisition module and the patent-related data obtained by the orthopedic implant patent data integration module.

[0095] The multi-source data feature construction module provides a unified data view that integrates all dimensions of features of the patent to be evaluated. Each piece of data can be logically linked through the patent ID or other associated identifiers to form a comprehensive feature vector that is ultimately used as input for the subsequent comprehensive value assessment model of orthopedic implant patents.

[0096] In a preferred embodiment of the present invention, the multi-source data features constructed by the multi-source data feature construction module are represented as D, then:

[0097]

[0098] in, For feature data that can represent the value of patented technology, For characteristic data that can represent the economic value of a patent, For feature data that can represent the legal value of a patent, ("c" stands for "clinic," and "r" stands for "research," representing characteristic data that can represent clinical and research aspects.) To represent the characteristic data of policies and regulations ("p" stands for "policies" and "s" stands for "supervision").

[0099] The comprehensive value assessment model for orthopedic implant patents is used to calculate the final value score of orthopedic implant patents. The model takes multi-source data features D obtained from the multi-source data feature construction module as input, and uses five sub-models to calculate evaluation data for five dimensions: clinical and research value, policy and regulatory value, patent legal value, patent technological value, and patent economic value. The outputs of the five sub-models are then fused using a gating network model.

[0100] The comprehensive value assessment model for orthopedic implant patents is represented by E. β (·), then we have:

[0101]

[0102] Among them, G β (·) represents the computational model, D tech (·) represents a sub-model used to obtain data for evaluating the value dimension of patented technologies. eco (·) represents a sub-model used to obtain data for evaluating the economic value dimension of patents, D legal (·) represents a sub-model used to obtain data for evaluating the legal value dimension of patents. c&r (·) represents the sub-model used to obtain evaluation data for clinical and research dimensions, D p &s (·) is a sub-model used to obtain evaluation data for policy and regulatory dimensions.

[0103] In E β (D) Model: Sub-model D tech (·) Based on the rules set in Table 2 below, the final patent technology value dimension evaluation data are obtained from the aspects of core innovation and patent technology depth, technology maturity (TRL), safety and effectiveness evidence, long-term stability / durability verification, multi-indication coverage depth, differentiation from existing technologies, and production quality control and scalability.

[0104] Table 2 Sub-model D tech Scoring criteria and data extraction methods

[0105]

[0106]

[0107] Based on Table 2 above, sub-model D tech (·) Obtain the final evaluation score D for the patent technology value dimension.tech The calculation process used is as follows:

[0108]

[0109] Where: the score for the secondary indicator is represented by S. i This indicates that i = 1, 2, 3, 4, ..., 7, which correspond to the scores of the seven secondary indicators in Table 2 above. Specifically: S1: Core Innovation and Patent Technology Depth Score; S2: Technology Maturity Level (TRL) Score; S3: Safety and Efficacy Evidence Score; S4: Long-Term Stability / Durability Verification Score; S5: Multi-Indication Coverage Depth Score; S6: Differentiation from Existing Technologies Score; S7: Production Quality Control and Scalability Capability Score; W i C is the weight of the score of the i-th secondary indicator; cor The correction coefficient is used to adjust the scores of the primary indicators to suit actual needs or further standardize the data. Each secondary indicator has a score S. i and weight W i All scores are given in Table 2 above. The score for each secondary indicator is obtained by comprehensively evaluating each sub-item, ranging from 0 to 10.

[0110] Sub-model D eco (·) Based on the rules set in Table 3 below, the final evaluation data of the patent economic value dimension is obtained from the aspects of potential market size, medical insurance / payment adaptability, centralized procurement adaptability, cost and price competitiveness, commercialization channel maturity, investment / financing attractiveness, and international market expansion potential.

[0111] Table 3 Sub-model D eco Scoring criteria and data extraction methods

[0112]

[0113]

[0114]

[0115] Based on Table 3 above, sub-model D eco (·) Obtain the final patent economic value dimension score D eco The calculation process used is as follows:

[0116]

[0117] Where: the score for the secondary indicator is represented by S. iThis indicates that i = 1, 2, 3, 4, ..., 6, corresponding to the scores of the six secondary indicators in Table 3 above. Specifically: S1: Medical insurance / payment suitability score, S2: centralized procurement suitability score, S3: cost and price competitiveness score, S4: commercialization channel maturity score, S5: investment / financing attractiveness score, and S6: international market expansion potential score; W i C is the weight of the score of the i-th secondary indicator; cor The correction coefficient is used to adjust the scores of the primary indicators to suit actual needs or further standardize the data. Each secondary indicator has a score S. i and weight W i All scores are given in Table 3 above. The score for each secondary indicator is obtained by comprehensively evaluating each sub-item, ranging from 0 to 10.

[0118] Sub-model D leagl (·) Based on the rules set in Table 4 below, the final evaluation data of the patent legal value dimensions are obtained from the aspects of patent family breadth, remaining protection period, claim coverage, patent legal status, citation influence, procedural security, freedom of implementation, and transferability / licensability potential.

[0119] Table 4 Sub-model D legal Scoring criteria and data extraction methods

[0120]

[0121]

[0122]

[0123] Based on Table 4 above, sub-model D legal (·) Obtain the final patent legal value dimension score D legal The calculation process used is as follows:

[0124]

[0125] Where: the score for the secondary indicator is represented by S. i This indicates that i = 1, 2, 3, 4, ..., 8, corresponding to the scores of the eight secondary indicators in Table 4 above. Specifically: S1: Patent family breadth score, S2: Remaining protection years score, S3: Claim coverage score, S4: Patent legal status score, S5: Citation influence score, S6: Procedural security score, S7: Freedom of implementation score, S8: Transferability / licensability potential score; W i C is the weight of the score of the i-th secondary indicator; cor The correction coefficient is used to adjust the scores of the primary indicators to suit actual needs or further standardize the data. Each secondary indicator has a score S.i and weight W i All scores are given in Table 4 above. The score for each secondary indicator is obtained by comprehensively evaluating each sub-item, ranging from 0 to 10.

[0126] Sub-model D c&r (·) Based on the rules set in Table 5 below, the final clinical and research dimension evaluation data were obtained from the following aspects: number of international guideline recommendations, number of domestic guideline recommendations, recommendation level, standard adoption rate, number of academic conference citations, academic visibility, mention rate in comments / meta-analysis, number of patent citations, indication matching degree, improvement in revision rate, achievement rate of prosthesis service life, decrease in complication rate, and improvement in PROMs.

[0127] Table 5 Sub-model D c&r Scoring criteria and data extraction methods

[0128]

[0129]

[0130]

[0131]

[0132] Based on Table 5 above, sub-model D c&r (·) Obtain the final clinical and research dimension score D c&r The calculation process used is as follows:

[0133]

[0134] Where: the score for the secondary indicator is represented by S. i Let i = 1, 2, 3, 4, ..., 13, corresponding to the scores of the 13 secondary indicators in Table 5 above. Specifically: S1: International guideline recommendation frequency score, S2: Domestic guideline recommendation frequency score, S3: Recommendation level score, S4: Standard adoption score, S5: Academic conference citation frequency score, S6: Academic visibility score, S7: Comment / Meta-analysis mention rate score, S8: Patent citation frequency score, S9: Indication matching score, S... 10 : Improvement score for renovation rate, S 11 : Prosthetic lifespan achievement rate score, S 12 : Complication rate decrease score, S 13 : Improvement score of PROMs; W i C is the weight of the score of the i-th secondary indicator; cor The correction coefficient is used to adjust the scores of the primary indicators to suit actual needs or further standardize the data. Each secondary indicator has a score S.i and weight W i All scores are given in Table 5 above. The score for each secondary indicator is obtained by comprehensively evaluating each sub-item, ranging from 0 to 10.

[0135] Sub-model D p&s (·) Based on the rules set in Table 6 below, the final policy and regulatory dimension evaluation data are obtained from aspects such as patient satisfaction, physician operation feedback, quality management system and business compliance, DRG / DIP grouping adaptation, post-marketing policy supervision, ethical review, adverse events and disputes, and innovation / priority review and approval and supporting measures.

[0136] Table 6 Sub-model D p&s Scoring criteria and data extraction methods

[0137]

[0138]

[0139] Based on Table 6 above, sub-model D p&s (·) Obtain the final policy and regulatory dimension score D p&s The calculation process used is as follows:

[0140]

[0141] Where: the score for the secondary indicator is represented by S. i This indicates that i = 1, 2, 3, 4, ..., 8, corresponding to the scores of the eight secondary indicators in Table 6 above. Specifically: S1: Patient satisfaction score, S2: Physician operation feedback score, S3: Quality management system and operational compliance score, S4: DRG / DIP grouping adaptation score, S5: Post-marketing policy supervision score, S6: Ethical review score, S7: Adverse events and disputes score, S8: Innovation / priority review and approval and supporting score; W i C is the weight of the score of the i-th secondary indicator; cor The correction coefficient is used to adjust the scores of the primary indicators to suit actual needs or further standardize the data. Each secondary indicator has a score S. i and weight W i All scores are given in Table 6 above. The score for each secondary indicator is obtained by comprehensively evaluating each sub-item, ranging from 0 to 10.

[0142] For the comprehensive value assessment model E of orthopedic implant patents β(·) To validate and evaluate the effectiveness and reliability of the comprehensive value score calculated by the model, an independent validation set can be constructed. For each patent evaluated in this validation set, multiple field experts can be invited to conduct independent subjective scoring or ranking to establish a "true" or "expert consensus" evaluation benchmark. Subsequently, the comprehensive value evaluation model E for orthopedic implant patents will be compared. β The performance of the model is evaluated by assessing the consistency between the calculated comprehensive value score and the expert scores or ranking results. Specifically, the Spearman correlation coefficient between the comprehensive value score output by the model and the expert scores is calculated to measure their consistency in ranking; at the same time, nonparametric rank correlation coefficients such as Kendall's Tau can also be used for auxiliary verification.

[0143] The evaluation and results output module is used to analyze, interpret, and visualize the results output by the comprehensive value assessment model for orthopedic implant patents, and finally generate an intuitive patent value assessment report for all patents to be evaluated.

[0144] In a preferred embodiment of the present invention, the evaluation and result output module further includes a value ranking and grade division unit, a result visualization unit, and a patent value analysis report generation unit.

[0145] The Value Ranking and Grading Unit is used to rank the comprehensive value scores of all patents to be evaluated from the comprehensive value assessment model for orthopedic implants by percentile (PR). The percentile ranking for each sample is calculated in descending order. Based on the percentile ranking, the Value Ranking and Grading Unit further categorizes all patents to be evaluated into different assessment levels, including: High Value (Top 20%): PRi ≥ 80%, Good Value (20%–50%): 50% ≤ PRi < 80%, Medium Value (50%–80%): 20% ≤ PRi < 50%, and Low Value (Bottom 20%): PRi < 20%.

[0146] The results visualization unit generates multi-dimensional patent value analysis charts based on the evaluation results output by the value ranking and grading unit and / or the comprehensive value assessment model for orthopedic implant patents. In a preferred embodiment of this invention, the patent value analysis charts may include: a radar chart: used to display the score distribution of a single patent across five dimensions: legal, technical, economic, clinical, and policy; a scatter plot / bubble chart: used to show the distribution of patents across two core dimensions (e.g., technological innovation vs. market potential), and can be overlaid with color or size to reflect the total value; a bar chart / line chart: used to display trend analysis results, such as the annual change in patent value in a specific technical field; and a heat map—used to display the patent value distribution among different technical subfields or applicants.

[0147] The patent value analysis report generation unit generates a systematic orthopedic implant patent value analysis report based on the evaluation results output by the value ranking and grading unit and / or the comprehensive value assessment model for orthopedic implant patents. In a preferred embodiment of this invention, the report includes: the purpose of the evaluation, an overview of the model methodology, key evaluation results (e.g., a list of high-value patents, the distribution of patent numbers at each grade), key insights (e.g., the technical characteristics of high-value patents, market trends), recommendations (e.g., technology layout, investment direction), and detailed charts and data support.

Claims

1. An orthopedic implant patent value assessment system, comprising: The method comprises the following steps: An orthopedic implant evaluation reference standard data acquisition module is used to acquire and structure key data indicators related to the patent to be evaluated from multi-source heterogeneous data, wherein the multi-source heterogeneous data is derived from orthopedic implant-related literature, clinical guidelines, and expert consensus; An orthopedic implant patent data integration module is used to efficiently and accurately collect patent-related data of the patent to be evaluated from multi-source heterogeneous data sources, and to perform deep cleaning, carding, and feature extraction; A multi-source data feature construction module is used to construct multi-source data features of the patent to be evaluated based on the key data indicators obtained by the orthopedic implant evaluation reference standard data acquisition module and the patent-related data obtained by the orthopedic implant patent data integration module, forming a comprehensive feature vector for input into the subsequent orthopedic implant patent comprehensive value evaluation model; An orthopedic implant patent comprehensive value evaluation model takes the multi-source data features D obtained by the multi-source data feature construction module as input, uses five sub-models to calculate the clinical and scientific research, policy and supervision, patent legal value, patent technical value, and patent economic value in five dimensions, respectively, calculates the dimension evaluation data, and then fuses the outputs of the five sub-models through a gating network model; An evaluation and result output module is used to analyze, interpret, and visualize the results output by the orthopedic implant patent comprehensive value evaluation model, and finally generate an intuitive patent value evaluation report for all patents to be evaluated.

2. The orthopedic implant patent value assessment system of claim 1, wherein, The key data indicators include family size / territorial coverage data, legal status / litigation risk data, claim coverage, patent citation network analysis data, core patent expiration date, international standard adoption data, industry technical specification compliance data, main endpoint improvement amplitude, 10-year revision rate, functional score change value, adverse event rate, complication reduction ratio, international guideline recommendation intensity data, indication coverage consistency data, CE / FDA / NMPA certification effectiveness data, label indication compliance data, national medical insurance directory inclusion data, payment proportion / self-pay amount, historical winning price trend data, procurement volume market share data, target patient penetration rate, unit price x dosage prediction model, BOM cost structure disassembly data, industry average gross profit comparison data, tertiary hospital coverage rate, channel cooperation stability data, satisfaction score, pain relief VAS score, operation convenience score, and intraoperative fitting problem incidence.

3. The orthopedic implant patent value assessment system of claim 1, wherein, The orthopedic implant evaluation reference standard data acquisition module systematically organizes the collected data, extracts text through regular expression and rule matching, or uses Doccano for manual annotation combined with natural language processing tools such as SpaCy for entity recognition and relationship extraction, and converts unstructured policy text into structured data.

4. The orthopedic implant patent value assessment system of claim 1, wherein, The orthopedic implant patent data integration module includes an orthopedic implant patent multi-element heterogeneous data acquisition unit and an orthopedic implant patent data cleaning and carding unit, wherein: The orthopedic implant patent multi-element heterogeneous data acquisition unit is used to acquire multi-source heterogeneous data indicators of all orthopedic implant field patents related to the patent to be evaluated; The orthopedic implant patent data cleaning and carding unit is used to clean and card the multi-source heterogeneous data indicators of the orthopedic implant field patents related to the patent to be evaluated. The orthopedic implant patent data cleaning and carding unit is used for deep cleaning, format conversion and structure integration of the original patent data obtained by the orthopedic implant patent multi-element isomer data collection unit.

5. An orthopaedic implant patent value assessment system as claimed in claim 4, wherein, The patent data indexes collected by the orthopedic implant patent multi-element isomer data collection unit include applicant, inventor, patent type, patent field, technical specification full text, disclosure number, priority, patent family information, legal status, citation relationship, transfer license record, litigation information, communication standard, designated national patent number, and expected expiration date.

6. The orthopedic implant patent value assessment system of claim 4, wherein, The data cleaning process adopted by the orthopedic implant patent data cleaning and carding unit specifically includes the following steps: Step 1, data retrieval and preliminary storage: the collected original patent data is preliminarily stored according to the field; Step 2, data exploration: descriptive statistical analysis is performed on the data to identify missing values, outliers, repeated items and inconsistent formats; Step 3, field cleaning: de-duplication, format unification and error value correction are performed on the key fields; Step 4, code and classification mapping: the classification codes in the original database are mapped to a unified classification system; Step 5, ontology and vocabulary standardization: for text fields such as patent specification and abstract, standardization and synonym normalization of professional vocabulary are performed; Step 6, structured text extraction: semi-structured or structured information is extracted from patent text using rule matching, regular expression or natural language processing technology.

7. The orthopedic implant patent value assessment system of claim 1, wherein, The multi-source data feature is represented as D, then: wherein, is characteristic data capable of representing the value of the patent technology, is characteristic data capable of representing the economic value of the patent, is characteristic data capable of representing the legal value of the patent, is characteristic data capable of representing the clinical and scientific research, is characteristic data capable of representing the policy and supervision.

8. The orthopedic implant patent value assessment system of claim 7, wherein, The orthopedic implant patent comprehensive value evaluation model is represented as E β (·), then: Wherein, G β (·) is a calculation model, D tech (·) is a sub-model for obtaining patent technology value dimension evaluation data, D eco (·) is a sub-model for obtaining patent economic value dimension evaluation data, D legal (·) is a sub-model for obtaining patent legal value dimension evaluation data, D c&r (·) is a sub-model for obtaining clinical and scientific research dimension evaluation data, D p&s (·) is a sub-model for obtaining policy and supervision dimension evaluation data.

9. The orthopedic implant patent value assessment system of claim 8, wherein, Sub-model D tech (·) Based on the rules set in the following table, the final patent technology value dimension evaluation data is obtained from the core innovation degree and the patent technology depth, technology maturity, safety and effectiveness evidence, long-term stability / durability verification, multi-indication coverage depth, differentiation from existing technology, production quality control and scale-up ability. Based on the above table, sub-model D tech (·) Obtain the final patent technology value dimension evaluation score D tech The calculation process used is: wherein: the secondary index score is represented by S i , wherein i = 1, 2, 3, 4, …, 7, respectively corresponding to the seven secondary index scores in the above table, S1: core innovation degree and patent technology depth score, S2: technology maturity score, S3: safety and effectiveness evidence score, S4: long-term stability / durability verification score, S5: multi-indication coverage depth score, S6: differentiation from existing technology score, S7: production quality control and scale-up capability score; W i is the weight of the ith secondary index score; C cor is the correction coefficient; Sub-model D eco (·) Based on the rules set in the table below, the final patent economic value dimension evaluation data is obtained from potential market size, medical insurance / payment adaptability, centralized procurement adaptability, cost and price competitiveness, commercialization channel maturity, investment / funding attractiveness, and international market expansion potential. Based on the above table, sub-model D eco (·) Obtain the final patent economic value dimension score D eco The calculation process used is: wherein: the secondary index score is represented by S i i = 1, 2, 3, 4, …, 6, respectively corresponding to the six secondary index scores in Table 3 above, S1: medical insurance / payment adaptability score, S2: collection purchase adaptability score, S3: cost and price competitiveness score, S4: commercialization channel maturity score, S5: investment / financing attractiveness score, S6: international market expansion potential score; Sub-model D legal (·) Based on the rules set in the following table, the final patent legal value dimension evaluation data is obtained from the patent family breadth, remaining protection period, claim coverage, patent legal status, citation influence, procedural security, freedom to operate space, and transferable / licensable potential. Based on the above table, sub-model D legal (·) Obtain the final patent legal value dimension score D legal The calculation process used is: wherein: the secondary indicator score is denoted by S i i = 1, 2, 3, 4, …, 8, corresponding to the eight secondary indicator scores in Table 4 above, Si: patent family breadth score, S2: remaining protection term score, S3: claim coverage score, S4: patent legal status score, S5: citation impact score, S6: procedural security score, S7: freedom to operate space score, S8: transferable / licensable potential score; Submodel D c&r (·) Based on the rules set in the following table, the final clinical and scientific dimension evaluation data is obtained from the international guideline recommended number of times, the domestic guideline recommended number of times, the recommended level, the standard adoption degree, the academic conference citation number, the academic visibility, the review / Meta analysis mention rate, the patent citation number, the indication matching degree, the revision rate improvement, the prosthesis use year limit achievement rate, the complication rate reduction, and the PROMs improvement range. Based on the above table, sub-model D c&r (·) Obtain final clinical and research dimension scores D c&r The calculation process used is: wherein: the secondary index score is represented by S i , i = 1, 2, 3, 4, …, 13, respectively corresponding to the 13 secondary index scores in the above table, S1: international guideline recommendation frequency score, S2: domestic guideline recommendation frequency score, S3: recommendation level score, S4: standard adoption degree score, S5: academic conference citation frequency score, S6: academic visibility score, S7: review / Meta analysis mention rate score, S8: patent citation frequency score, S9: indication matching degree score, S 1@ : revision rate improvement score, S 11 : prosthesis use duration achievement rate score, S 12 : complication rate reduction score, S 13 : PROMs improvement amplitude score; Sub-model D p&s (·) Based on the rules set in the table below, the final policy and regulatory dimension evaluation data is obtained from patient satisfaction, physician operation feedback, quality management system and business compliance, DRG / DIP grouping adaptation, post-market policy supervision, ethical review, adverse events and disputes, innovation / priority review approval and supporting aspects; Based on the above table, sub-model D p&s (·) Obtain the final policy and regulatory dimension score D p&s The calculation process used is: wherein: the secondary index score is represented by S i i = 1, 2, 3, 4, …, 8, respectively corresponding to the eight secondary index scores in the above table, S1: patient satisfaction score, S2: physician operation feedback score, S3: quality management system and business compliance score, S4: DRG / DIP grouping adaptation score, S5: post-marketing policy supervision score, S6: ethical review score, S7: adverse events and dispute score, S8: innovation / priority review approval and supporting score.

10. The orthopedic implant patent value assessment system of claim 1, wherein, The evaluation and result output module includes value ranking and grade division unit, result visualization unit and patent value analysis report generation unit, wherein: The value ranking and grade division unit is used for ranking the comprehensive value score results of all patents to be evaluated output by the orthopedic implant patent comprehensive value evaluation model by percentile, and calculating the percentile ranking of each sample in descending order; The result visualization unit generates multi-dimensional patent value analysis charts according to the evaluation results output by the value ranking and grade division unit and / or the orthopedic implant patent comprehensive value evaluation model; The patent value analysis report generation unit forms a systematic orthopedic implant patent value analysis report according to the evaluation results output by the value ranking and grade division unit and / or the orthopedic implant patent comprehensive value evaluation model.

Images