Orthopedic related drug patent value evaluation system

Abstract

The technical scheme of the invention discloses an orthopedics-related drug patent value evaluation system, which is characterized by comprising an orthopedics-related drug evaluation reference standard data acquisition module; an orthopedic related drug patent data integration module; a multi-source data feature construction module; an orthopedic related drug patent comprehensive value evaluation model; and an evaluation and result output module. According to the system disclosed by the invention, traditional patent evaluation indexes such as law, economy and technology can be combined with clinical effects, scientific research, medicine and health policies, supervision systems and the like of orthopedic drugs, so that dynamic evaluation and prospective prediction of the patent values of the orthopedic related drugs are realized. Meanwhile, multi-source data acquisition and an advanced machine learning algorithm are introduced, data scientificity and rationality are improved by using multiple model methods, an evaluation system exclusive to the orthopedics-related pharmaceutical industry is constructed, a closed-loop process of achievement transformation-patent value evaluation is formed, and scientificity, adaptability and practicability of evaluation results are ensured.

Description

Technical Field

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

[0002] Drug treatment for orthopedic diseases encompasses multiple aspects, including analgesia and anti-inflammation, osteoporosis prevention and treatment, and bone repair. Commonly used drugs include traditional Chinese medicine, Western medicine, and biologics. Some mild cases can be improved through conservative treatment, but for patients with severe osteoporosis, intractable bone pain, or insufficient bone healing capacity, systemic drug therapy is necessary to improve the bone metabolic environment and alleviate symptoms, thereby supporting bone healing and functional recovery. There are numerous types of orthopedic drugs, including traditional small-molecule drugs (such as nonsteroidal anti-inflammatory drugs and bisphosphonates), novel biologics (such as RANKL recombinant protein inhibitors and anabolic catabolites), and some traditional Chinese medicines widely used in clinical practice. In recent years, the number of patents for orthopedic drugs has increased rapidly, but the quality and translational effectiveness vary significantly. This makes identifying high-value, translatable, and investable orthopedic drug patents and establishing scientific and systematic evaluation methods an urgent problem to be solved.

[0003] Current patent valuation systems primarily extract indicators from three dimensions: legal, technical, and economic, combining qualitative and quantitative tools for analysis. Common methods include three asset valuation theories: the cost approach, the market approach, and the income approach. These methods treat patents as intangible assets and quantify them in monetary terms. Specifically, the cost approach emphasizes the replacement cost of the patent-related technology, the market approach relies on comparable transaction cases, and the income approach calculates patent value by discounting future expected revenue. Existing research institutions (such as universities, hospitals, and government departments) focus more on constructing indicator systems, asset valuation companies emphasize assetization and revenue prediction, and patent agencies emphasize full lifecycle management of patents. However, the current evaluation system still has several shortcomings: First, the evaluation indicators are highly subjective, difficult to fully quantify, and lack specific clinical and policy dimensions for orthopedic drugs; second, it lacks industry universality, making it difficult to achieve uniform application across different drug types; third, data resources are limited, especially in PK / PD, real-world clinical evidence, and medical insurance payment databases, which are still incomplete, lacking real feedback on clinical efficacy and patient follow-up data, thus restricting the accuracy and breadth of the evaluation system.

[0004] In the patent evaluation process for orthopedic drugs, special attention needs to be paid to field-specific influencing factors. On the one hand, patent value depends not only on technological innovation but also on factors throughout the entire lifecycle, including clinical trial results, real-world studies, drug registration and approval, and post-marketing surveillance. On the other hand, technical indicators such as molecular innovation, specific mechanism of action, improved delivery systems, pharmacokinetic / pharmacodynamic (PK / PD) and bone tissue exposure should be combined with economic and policy factors such as regulatory policies, medical insurance access, volume-based procurement, and hospital bidding. Simultaneously, key elements such as the drug's applicability in clinical pathways, its synergistic use with implants and surgeries, patient compliance, and adoption of clinical guidelines should also be considered. 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, follow-up results, real-world evidence and other clauses.

[0007] (2) The existing patent evaluation indicators are not broad enough, and lack quantitative evaluation indicators in terms of life cycle management, demand satisfaction, capital attractiveness, indication expansion, quality and process reliability of orthopedic drugs.

[0008] (3) The existing evaluation index system for orthopedic drug patents is not advanced enough. It cannot integrate multi-source data from home and abroad and incorporate advanced machine learning methods. The processing efficiency is slow and the accuracy of the results is low.

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

[0010] The data acquisition module for the evaluation reference standards of orthopedic drugs 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 literature, clinical guidelines and expert consensus on orthopedic drugs.

[0011] The orthopedic drug 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 drug evaluation reference standard data acquisition module and the patent-related data obtained by the orthopedic drug patent data integration module, forming a comprehensive feature vector that is finally used as input for the subsequent comprehensive value assessment model of orthopedic drug patents.

[0013] The comprehensive value assessment model for orthopedic drug 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 drug 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 data acquisition module for the evaluation reference standard of orthopedic drugs 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 drug patent data integration module includes a multi-dimensional heterogeneous data acquisition unit for orthopedic drug patents and a data cleaning and sorting unit for orthopedic drug patents, wherein:

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

[0019] The orthopedic drug 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 drug patent multivariate heterogeneous data acquisition unit.

[0020] Preferably, the patent data indicators collected by the orthopedic-related drug 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 / licensing record, litigation information, communication standard, designated country patent number, and expected expiration date.

[0021] Preferably, the data cleaning process adopted by the orthopedic drug 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-related drug patents is represented by E. + (·), then we have:

[0032]

[0033] Among them, G+ (·) represents the computational model, D (e, -(·) represents a sub-model used to obtain data for evaluating the value dimension of patented technologies, D e,0 (·) represents a sub-model used to obtain data for evaluating the economic value dimension of patents, D .e / a. (·) represents a sub-model used to obtain data for evaluating the legal value dimension of patents. ,&2 (·) represents the sub-model used to obtain evaluation data for clinical and research dimensions, D p&3 (·) is a sub-model used to obtain evaluation data for policy and regulatory dimensions.

[0034] Preferably, sub-model D (e, -(·)Based on the rules set in the table below, the final patent technology value dimension evaluation data are obtained from the aspects of molecular innovation, mechanism of action and pathway specificity, formulation and delivery system, PK / PD and bone tissue exposure, quality and process controllability, scientific research impact and knowledge diffusion, translational readiness, and coherence with bone biomarkers.

[0035]

[0036] Based on the table above, sub-model D (e,- (·) Obtain the final evaluation score D for the patent technology value dimension. (e,- The calculation process used is as follows:

[0037]

[0038] Wherein: the secondary indicator score is represented by S5, where i = 1, 2, 3, 4, ..., 8, corresponding to the 8 secondary indicator scores in the table above, S8: molecular innovation score, S9: mechanism of action and pathway specificity score, S... : : Formulation and delivery system score, S ； PK / PD and bone tissue exposure score, S < Quality and process controllability score, S ＝ Research impact and knowledge diffusion score, S > : Transformation readiness score, S6: coherence score with bone biomarkers; W5 is the weight of the i-th secondary indicator score; C, 02 For correction factors;

[0039] Sub-model D e,0 (·) Based on the rules set in the table below, the final patent economic value dimension evaluation data is obtained from the aspects of target market size and growth, unmet needs, competitive landscape and generic / biosimilar drug risks, health insurance / payment accessibility, pricing and cost structure, life cycle and indication expansion, cooperation and capital attractiveness, and real-world penetration.

[0040]

[0041] Based on the table above, sub-model D e,0 (·) Obtain the final patent economic value dimension score D e,0 The calculation process used is as follows:

[0042]

[0043] Wherein: the scores of the secondary indicators are represented by S5, i = 1, 2, 3, 4, ..., 8, corresponding to the scores of the 8 secondary indicators in the table above, S8: target market size and growth score, S9: unmet needs score, S... : Competitive landscape and generic / biosimilar drug risk score, S ； : Health insurance / payment accessibility score, S < Pricing and cost structure score, S ＝ Lifecycle and indication expansion score, S > S6: Score for cooperation and capital attraction; S6: Score for real-world penetration.

[0044] Sub-model D .e / a. (·) Based on the rules set in the table below, the final evaluation data of patent legal value dimensions are obtained from the aspects of patent family size and geographical coverage, remaining protection period and PTE / compensation, claim coverage, data exclusivity / patent linkage, citation influence, FTO and infringement risk, litigation / opposition / invalidation history, and life cycle management.

[0045]

[0046] Based on the table above, sub-model D .e / a. (·) Obtain the final patent legal value dimension score D .e / a. The calculation process used is as follows:

[0047]

[0048] Wherein: the secondary indicator score is represented by S5, i = 1, 2, 3, 4, ..., 8, corresponding to the 8 secondary indicator scores in the table above, S8: patent family size and geographic coverage score, S9: remaining protection period and PTE / compensation score, S... : Claim coverage score, S ； Data exclusivity / patent link score, S < Citation influence score, S ＝ FTO and Infringement Risk Score, S > S6: Litigation / Objection / Invalidity History Score;

[0049] Sub-model D ,&2 (·) Based on the rules set in the table below, the final clinical and research dimension evaluation data were obtained from the following aspects: clinical stage and endpoint achievement, analgesia and anti-inflammatory evidence, swelling reduction / microcirculation, osteogenic / bone repair endpoint, neuroprotection / regeneration, safety and tolerability, real-world evidence, guidelines / consensus and academic impact, and imaging / functional composite endpoint.

[0050]

[0051] Based on the table above, sub-model D ,&2 (·) Obtain the final score D for the clinical and research dimensions. ,&2 The calculation process used is as follows:

[0052]

[0053] Wherein: the scores of the secondary indicators are represented by S5, i = 1, 2, 3, 4, ..., 9, corresponding to the scores of the 9 secondary indicators in the table above, S8: score for clinical stage and endpoint achievement, S9: score for evidence of analgesia and anti-inflammation, S... : Swelling reduction / microcirculation score, S ； Osteogenesis / bone repair endpoint score, S < Neuroprotection / regeneration score, S ＝ Safety and tolerability score, S > S6: Real-world evidence score; S7: Guidelines / consensus and academic impact score; S8: Image / functional composite endpoint score.

[0054] Sub-model D p&3 (·) Based on the rules set in the table below, the final policy and regulatory dimension evaluation data will be obtained from the aspects of review channels and qualifications, data exclusivity and market protection, patent linkage / early dispute resolution, GMP / pharmacovigilance / postmarket surveillance, ICH and international harmonization, ethics and data compliance, national and local policy incentives, standard and guideline adoption, and risk events and compliance records.

[0055]

[0056] Based on the table above, sub-model D p&3 (·) Obtain the final policy and regulatory dimension score D p&3 The calculation process used is as follows:

[0057]

[0058] Wherein: the scores of the secondary indicators are represented by S5, i = 1, 2, 3, 4, ..., 9, corresponding to the scores of the 9 secondary indicators in the table above, S8: review channel and qualification score, S9: data exclusivity and market protection score, S... :Patent Link / Early Dispute Resolution Score, S ； GMP / Pharmacovigilance / Postmarket Surveillance Score, S < ICH and international harmonization score, S ＝ Ethics and data compliance score, S > S6: National and local policy incentives score; S7: Standards and guidelines adoption score; S8: Risk events and compliance record score.

[0059] 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:

[0060] 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 drug patents in percentile order, and calculate the percentile ranking of each sample in descending order.

[0061] 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-related drug patents.

[0062] The patent value analysis report generation unit generates a systematic patent value analysis report for orthopedic-related drugs based on the evaluation results output by the value ranking and grading unit and / or the comprehensive value assessment model for orthopedic-related drug patents.

[0063] The system disclosed in this invention combines traditional patent evaluation indicators such as legal, economic, and technical factors with the clinical efficacy, scientific research, medical and health policies, and regulatory systems of orthopedic drugs, thereby achieving dynamic evaluation and forward-looking prediction of the patent value of orthopedic-related drugs. Simultaneously, this invention introduces multi-source data collection and advanced machine learning algorithms, utilizing various model methods to enhance the scientific validity and rationality of the data, constructing an industry-specific evaluation system for orthopedic-related drugs, and forming a closed-loop process of "technology transfer—patent value assessment," ensuring the scientific validity, adaptability, and practicality of the evaluation results.

[0064] The value assessment system disclosed in this invention can comprehensively present the value composition of orthopedic drug patents, covering multiple dimensions such as molecular innovation, specific mechanism of action, formulation and delivery system, PK / PD and bone tissue exposure, quality and process controllability, legal stability and market application potential, so as to achieve a multi-faceted and multi-angle systematic evaluation of orthopedic drug patents.

[0065] The system disclosed in this invention can not only help pharmaceutical investors identify and distinguish high-value patents, providing a scientific basis for investment, financing and technology transfer, but also provide inspiration for entrepreneurs who intend to carry out orthopedic drug research and development and patent layout, revealing what constitutes a high-value drug patent and its potential for transformation, thereby promoting the research, screening and industrial application of more innovative drugs and accelerating the efficient transformation of orthopedic drug scientific and technological achievements. Attached Figure Description

[0066] Figure 1 The working process of the system disclosed in this invention is illustrated. Detailed Implementation

[0067] 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.

[0068] like Figure 1 As shown in the figure, the patent value assessment system for orthopedic related drugs disclosed in this embodiment of the invention includes a data acquisition module for orthopedic related drug evaluation reference standards, a patent data integration module for orthopedic related drugs, a multi-source data feature construction module, a comprehensive value assessment model for orthopedic related drug patents, and an evaluation and result output module.

[0069] The data acquisition module for the evaluation reference standards of orthopedic drugs is used to acquire and structure key data indicators from multi-source heterogeneous data related to orthopedic drugs, including relevant literature, clinical guidelines, and expert consensus, to reflect the guiding significance and application potential of the patent technology in clinical practice.

[0070] In one preferred embodiment of the present invention, the key data indicators acquired by the orthopedic drug evaluation reference standard data acquisition module are shown in Table 1 below.

[0071] Table 1. Reference Standards for Evaluation of Orthopedic Related Drugs

[0072]

[0073] 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) / De Novo / 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.

[0074] The data acquisition module for orthopedic drug evaluation reference standards 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 texts into structured data.

[0075] The orthopedic drug 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 drug patent data integration module further includes an orthopedic drug patent multi-source heterogeneous data acquisition unit and an orthopedic drug patent data cleaning and sorting unit.

[0076] The orthopedic drug-related patent multi-source heterogeneous data acquisition unit is used to acquire multi-source heterogeneous data indicators of all orthopedic drug-related patents related to the patent to be evaluated. In one preferred embodiment of the invention, the orthopedic drug-related patent multi-source heterogeneous data acquisition unit primarily uses API interfaces to batch-screen and collect series of orthopedic drug-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 drug-related 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.

[0077] The patent data indicators collected by the multi-heterogeneous data acquisition unit for orthopedic drug patents 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, and expected expiration date.

[0078] The orthopedic drug 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 drug patent multivariate heterogeneous data acquisition unit, to ensure the high quality and usability of the data, and to provide standardized input for subsequent feature extraction.

[0079] In a preferred embodiment of the present invention, the data cleaning process employed by the orthopedic drug patent data cleaning and sorting unit specifically includes the following steps:

[0080] Step 1, Data Retrieval and Preliminary Storage: The collected original patent data is initially stored according to fields;

[0081] Step 2, Data Exploration: Perform descriptive statistical analysis on the data to identify missing values, outliers, duplicates, and inconsistent formats;

[0082] 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;

[0083] Step 4, Code and Classification Mapping: Map the classification codes (such as IPC classifications) in the original database to a unified classification system;

[0084] Step 5, Ontology and Lexical Standardization: Standardize professional terms and normalize synonyms for text fields such as patent specifications and abstracts;

[0085] 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).

[0086] 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 drug evaluation reference standard data acquisition module and the patent-related data obtained by the orthopedic drug patent data integration module.

[0087] 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 data item 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 drug patents.

[0088] 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:

[0089]

[0090] 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").

[0091] The comprehensive value assessment model for orthopedic drug patents is used to calculate the final value score of orthopedic drug 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 the evaluation data for each of the five dimensions: clinical and research value, policy and regulatory value, patent legal value, patent technological value, and patent economic value. Finally, a gating network model is used to fuse the outputs of the five sub-models.

[0092] The comprehensive value assessment model for orthopedic drug patents is represented by e. + (·), then we have:

[0093]

[0094] Among them, G + (·) represents the computational model, D (e,- (·) represents a sub-model used to obtain data for evaluating the value dimension of patented technologies. e,0 (·) represents a sub-model used to obtain data for evaluating the economic value dimension of patents, D .e / a. (·) represents a sub-model used to obtain data for evaluating the legal value dimension of patents. ,&2 (·) represents the sub-model used to obtain evaluation data for clinical and research dimensions, D p&3 (·) is a sub-model used to obtain evaluation data for policy and regulatory dimensions.

[0095] In E + (D) Model: Sub-model D (e,- (·) Based on the rules set in Table 2 below, the final patent technology value dimension evaluation data are obtained from the aspects of molecular innovation, mechanism of action and pathway specificity, formulation and delivery system, PK / PD and bone tissue exposure, quality and process controllability, scientific research impact and knowledge diffusion, translational readiness, and coherence with bone biomarkers.

[0096] Table 2 Sub-model D (e,- Scoring criteria and data extraction methods

[0097]

[0098] Based on Table 2 above, sub-model D (e,- (·) Obtain the final evaluation score D for the patent technology value dimension. (e,- The calculation process used is as follows:

[0099]

[0100] Wherein: the secondary indicator scores are represented by S5, where i = 1, 2, 3, 4, ..., 8, corresponding to the 8 secondary indicator scores in Table 2 above. Specifically—S8: molecular innovation score, S9: mechanism of action and pathway specificity score, S... : : Formulation and delivery system score, S ； PK / PD and bone tissue exposure score, S < Quality and process controllability score, S ＝ Research impact and knowledge diffusion score, S > : Transformation readiness score, S6: coherence score with bone biomarkers; W5 is the weight of the i-th secondary indicator score; C, 02 The correction coefficient is used to adjust the scores of the primary indicators to suit actual needs or further standardize the data. The score S5 and weight W5 for each secondary indicator are given in Table 2 above. The score of each secondary indicator is obtained by comprehensively evaluating each sub-item, ranging from 0 to 10.

[0101] Sub-model D e,0 (·) Based on the rules set in Table 3 below, the final patent economic value dimension evaluation data is obtained from the aspects of target market size and growth, unmet needs, competitive landscape and generic / biosimilar drug risks, health insurance / payment accessibility, pricing and cost structure, life cycle and indication expansion, cooperation and capital attractiveness, and real-world penetration.

[0102] Table 3 Sub-model D e,0 Scoring criteria and data extraction methods

[0103]

[0104] Based on Table 3 above, sub-model D e,0 (·) Obtain the final patent economic value dimension score D e,0 The calculation process used is as follows:

[0105]

[0106] Wherein: the scores of the secondary indicators are represented by S5, i = 1, 2, 3, 4, ..., 8, corresponding to the scores of the 8 secondary indicators in Table 3 above. Specifically—S8: target market size and growth score, S9: unmet needs score, S... : Competitive landscape and generic / biosimilar drug risk score, S ； : Health insurance / payment accessibility score, S < Pricing and cost structure score, S ＝ Lifecycle and indication expansion score, S > S6: Score for cooperation and capital attraction; W5: Weight of the i-th secondary indicator score; C, 02 The correction coefficient is used to adjust the scores of the primary indicators to suit actual needs or to further standardize the data. The score S5 and weight W5 for each secondary indicator are given in Table 3 above. The score of each secondary indicator is obtained by comprehensively evaluating each sub-item, ranging from 0 to 10.

[0107] Sub-model D .e / a. (·) Based on the rules set in Table 4 below, the final evaluation data of patent legal value dimensions are obtained from the aspects of patent family size and geographical coverage, remaining protection period and PTE / compensation, claim coverage, data exclusivity / patent linkage, citation influence, FTO and infringement risk, litigation / opposition / invalidation history, and life cycle management.

[0108] Table 4 Sub-model D .e / a. Scoring criteria and data extraction methods

[0109]

[0110] Based on Table 4 above, sub-model D .e / a. (·) Obtain the final patent legal value dimension score D .e / a. The calculation process used is as follows:

[0111]

[0112] Wherein: the secondary indicator scores are represented by S5, i = 1, 2, 3, 4, ..., 8, corresponding to the 8 secondary indicator scores in Table 4 above, specifically—S8: Patent family size and geographic coverage score, S9: Remaining protection period and PTE / compensation score, S... : Claim coverage score, S ； Data exclusivity / patent link score, S < Citation influence score, S ＝ FTO and Infringement Risk Score, S > S6: Historical scores for litigation / objections / invalidity; W5: Weight of the i-th secondary indicator score; C, 02 The correction coefficient is used to adjust the scores of the primary indicators to suit actual needs or further standardize the data. The score S5 and weight W5 for each secondary indicator are given in Table 4 above. The score of each secondary indicator is obtained by comprehensively evaluating each sub-item, ranging from 0 to 10.

[0113] Sub-model D ,&2 (·) Based on the rules set in Table 5 below, the final clinical and research dimension evaluation data were obtained from the following aspects: clinical stage and endpoint achievement, analgesia and anti-inflammatory evidence, swelling reduction / microcirculation, osteogenic / bone repair endpoint, neuroprotection / regeneration, safety and tolerability, real-world evidence, guidelines / consensus and academic impact, and imaging / functional composite endpoint.

[0114] Table 5 Sub-model D ,&2 Scoring criteria and data extraction methods

[0115]

[0116] Based on Table 5 above, sub-model D ,&2 (·) Obtain the final score D for the clinical and research dimensions. ,&2 The calculation process used is as follows:

[0117]

[0118] Wherein: the scores of the secondary indicators are represented by S5, i = 1, 2, 3, 4, ..., 9, corresponding to the scores of the 9 secondary indicators in Table 5 above. Specifically—S8: Clinical stage and endpoint achievement score, S9: Analgesia and anti-inflammatory evidence score, S... : Swelling reduction / microcirculation score, S ； Osteogenesis / bone repair endpoint score, S < Neuroprotection / regeneration score, S ＝ Safety and tolerability score, S >S6: Real-world evidence score; S7: Guideline / Consensus and Academic Impact score; S8: Image / Functional Composite Endpoint score; W9: Weight of the i-th secondary indicator score; C, 02 The correction coefficient is used to adjust the scores of the primary indicators to suit actual needs or to further standardize the data. The score S5 and weight W5 for each secondary indicator are given in Table 5 above. The score of each secondary indicator is obtained by comprehensively evaluating each sub-item, ranging from 0 to 10.

[0119] Sub-model D p&3 (·) Based on the rules set in Table 6 below, the final policy and regulatory dimension evaluation data are obtained from the aspects of review channels and qualifications, data exclusivity and market protection, patent linkage / early dispute resolution, GMP / pharmacovigilance / postmarket surveillance, ICH and international harmonization, ethics and data compliance, national and local policy incentives, standard and guideline adoption, and risk events and compliance records.

[0120] Table 6 Sub-model D p&3 Scoring criteria and data extraction methods

[0121]

[0122] Based on Table 6 above, sub-model D p&3 (·) Obtain the final policy and regulatory dimension score D p&3 The calculation process used is as follows:

[0123]

[0124] Wherein: the scores of the secondary indicators are represented by S5, i = 1, 2, 3, 4, ..., 9, corresponding to the scores of the 9 secondary indicators in Table 6 above, specifically—S8: review channel and qualification score, S9: data exclusivity and market protection score, S... : Patent Link / Early Dispute Resolution Score, S ； GMP / Pharmacovigilance / Postmarket Surveillance Score, S < ICH and international harmonization score, S ＝ Ethics and data compliance score, S > S6: National and local policy incentive score; S7: Standards and guidelines adoption score; S8: Risk events and compliance record score; W9: Weight of the i-th secondary indicator score; C, 02 The correction coefficient is used to adjust the scores of the primary indicators to suit actual needs or further standardize the data. The score S5 and weight W5 for each secondary indicator are given in Table 6 above. The score of each secondary indicator is obtained by comprehensively evaluating each sub-item, ranging from 0 to 10.

[0125] E, a comprehensive value assessment model for orthopedic drug patents. + (·) Validation is conducted to assess the validity and reliability of the comprehensive value score calculated by the model. To this end, an independent validation set can be constructed, and for each patent evaluated in this 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-related drug patents will be compared. + 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; alternatively, nonparametric rank correlation coefficients such as Kendall's Tau can be used for auxiliary verification.

[0126] The evaluation and results output module is used to analyze, interpret, and visualize the results output by the comprehensive value assessment model for orthopedic drug patents, and finally generate an intuitive patent value assessment report for all patents to be evaluated.

[0127] 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.

[0128] 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 drug patents using percentile ranking (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%.

[0129] 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-related drug 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, technological, 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 technological field; and a heat map—used to display the patent value distribution among different technological subfields or applicants.

[0130] The patent value analysis report generation unit generates a systematic patent value analysis report for orthopedic-related drugs based on the evaluation results output by the value ranking and grading unit and / or the comprehensive value assessment model for orthopedic-related drug 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 drug-related patent value evaluation system characterized by, The application relates to a bone surgery related drug patent comprehensive value evaluation method and system. The key data indexes include the same family size / region coverage data, legal state / litigation risk data, patent 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, function score change value, adverse event occurrence 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-payment amount, historical bid price trend data, procurement quantity market share data, target patient penetration rate, unit price*quantity 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 adaptation problem occurrence rate. The bone surgery related drug patent data integration module includes a bone surgery related drug patent multi-element heterogeneous data acquisition unit and a bone surgery related drug patent data cleaning and carding unit, wherein: The bone surgery related drug patent data integration module includes a bone surgery related drug patent multi-element heterogeneous data acquisition unit and a bone surgery related drug patent data cleaning and carding unit, wherein: ​ ​ 2. The orthopedic-related drug patent value assessment system of claim 1, wherein, ​ 3. The orthopedic related drug patent value assessment system of claim 1, wherein, ​ 4. The orthopedic drug patent value assessment system of claim 1, wherein, ​ The orthopedics-related drug patent multi-source heterogeneous data collection unit is configured to collect all multi-source heterogeneous data indicators of patents in the field of orthopedics-related drugs. The orthopedics-related drug patent data cleaning and carding unit is configured to clean, format convert and structure integrate the original patent data collected by the orthopedics-related drug patent multi-source heterogeneous data collection unit.

5. The orthopedic related drug patent value assessment system of claim 4, wherein, The patent data collected by the orthopedics-related drug patent multi-source heterogeneous data collection unit includes the following indicators: applicant, inventor, patent type, patent field, full text of technical specification, publication number, priority, patent family information, legal status, citation relationship, transfer and licensing record, litigation information, communication standard, designated national patent number, and expected expiration date.

6. The orthopedic related drug patent value assessment system of claim 4, wherein, The data cleaning process adopted by the orthopedics-related drug patent data cleaning and carding unit includes the following steps: Step 1: Data retrieval and preliminary storage: the collected original patent data is preliminarily stored by field; Step 2: Data exploration: descriptive statistical analysis is performed on the data to identify missing values, outliers, duplicates and inconsistent formats; Step 3: Field cleaning: key fields are de-duplicated, formatted and error values are corrected; Step 4: Code and classification mapping: mapping the classification codes in the original database to a unified classification system; Step 5: Ontology and vocabulary standardization: standardizing professional vocabulary and synonym normalization for text fields such as patent specification and abstract; Step 6: Structured text extraction: using rule matching, regular expressions or natural language processing techniques to extract semi-structured or structured information from patent text.

7. The orthopedic related drug patent value assessment system of claim 1, wherein, The multi-source data feature is represented as D, and the following equation holds: 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 related drug patent value assessment system of claim 7, wherein, The orthopedic related drug patent comprehensive value evaluation model is represented as E + (·), then: Wherein, G + (·) is a calculation model, D (e,- (·) is a sub-model for obtaining patent technology value dimension evaluation data, D e,0 (·) is a sub-model for obtaining patent economic value dimension evaluation data, D .e / a. (·) is a sub-model for obtaining patent legal value dimension evaluation data, D ,&2 (·) is a sub-model for obtaining clinical and scientific research dimension evaluation data, D p&3 (·) is a sub-model for obtaining policy and supervision dimension evaluation data.

9. The orthopedic related drug patent value assessment system of claim 8, wherein, Submodel D (e, -(·) Based on the rules set in the table below, the final patent technology value dimension evaluation data is obtained from the aspects of molecular innovation, mechanism and pathway specificity, preparation and delivery system, PK / PD and bone tissue exposure, quality and process controllability, scientific research influence and knowledge diffusion, transformation readiness, and consistency with bone biomarkers; Based on the above table, sub-model D (e,- (·) Obtain the final patent technology value dimension evaluation score D (e,- The calculation process used is: wherein: the secondary index score is denoted by S5, wherein i = 1, 2, 3, 4, …, 8, respectively corresponding to the eight secondary index scores in the above table, S8: molecular innovativeness score, S9: mechanism of action and pathway specificity score, S : : formulation and delivery system score, S ； : PK / PD and bone tissue exposure score, S < : quality and process controllability score, S ＝ : scientific impact and knowledge diffusion score, S > : translational readiness score, S6: coherence with bone biomarkers score; W5 is the weight of the i-th secondary index score; C, 02 is the correction coefficient; Submodel D e,0 (·) Based on the rules set in the table below, the final patent economic value dimension evaluation data is obtained from the target market size and growth, unmet demand, competition pattern and imitation / biosimilar risk, medical insurance / payment accessibility, pricing and cost structure, life cycle and indication expansion, cooperation and capital attraction, real world penetration Based on the above table, sub-model D e,0 (·) Obtain the final patent economic value dimension score D e,0 The calculation process used is: wherein: the secondary indicator score is represented by S5, i = 1, 2, 3, 4, …, 8, corresponding to the eight secondary indicator scores in the table above, S8: target market size and growth score, S9: unmet need score, S : : competition landscape and generic / biosimilar risk score, S ； : insurance / payment accessibility score, S < : pricing and cost structure score, S ＝ : life cycle and indication expansion score, S > : partnership and capital attractiveness score, S6: real-world penetration score; Sub-model D .e / a. (·) Based on the rules set in the following table, the final patent legal value dimension evaluation data is obtained from the aspects of patent family size and regional coverage, remaining protection period and PTE / compensation, claim coverage, data monopoly / patent linkage, citation influence, FTO and infringement risk, litigation / objection / invalidation history, and life cycle management. Based on the above table, sub-model D .e / a. (·) Obtain the final patent legal value dimension score D .e / a. The calculation process used is: wherein: the secondary indicator score is represented by S5, i = 1, 2, 3, 4, …, 8, corresponding to the eight secondary indicator scores in the table above, S8: patent family size and geographical coverage score, S9: remaining protection period and PTE / compensation score, S : : claim coverage score, S ； : data exclusivity / patent linkage score, S < : citation impact score, S ＝ : FTO and infringement risk score, S > : litigation / office action history score, S6: life cycle management score; Submodel D ,&2 (·) Based on the rules set in the table below, the final clinical and scientific dimension evaluation data is obtained from the clinical stage and endpoint, analgesic and anti-inflammatory evidence, detumescence / microcirculation, osteogenesis / bone repair endpoints, neuroprotection / regeneration, safety and tolerability, real-world evidence, guidelines / consensus and academic impact, image / function composite endpoints. Based on the above table, sub-model D ,&2 (·) Obtain final clinical and research dimension scores D ,&2 The calculation process used is: wherein: the secondary indicator score is represented by S5, i = 1, 2, 3, 4, …, 9, corresponding to the 9 secondary indicator scores in the table above, S8: clinical stage and endpoint achievement score, S9: analgesia and anti-inflammatory evidence score, S : : decongestion / microcirculation score, S ； : osteogenesis / bone repair endpoint score, S < : neuroprotection / regeneration score, S ＝ : safety and tolerability score, S > : real world evidence score, S6: guideline / consensus and academic impact score, S9: imaging / function composite endpoint score; Submodel D p&3 (·) Based on the rules set in the table below, the final policy and regulatory dimension evaluation data are obtained from the review channels and qualifications, data exclusivity and market protection, patent linkage / early dispute resolution, GMP / drug safety / post-market monitoring, ICH and international coordination, ethics and data compliance, national and local policy incentives, standard and guideline adoption, risk events and compliance records. Based on the above table, sub-model D p&3 (·) Obtain the final policy and regulatory dimension score D p&3 The calculation process used is: wherein: the secondary indicator score is represented by S5, i = 1, 2, 3, 4, 9, corresponding to the nine secondary indicator scores in the table above, S8: review pathway and eligibility score, S9: data exclusivity and market protection score, S : : patent linkage / early dispute resolution score, S ； : GMP / drug vigilance / post-market surveillance score, S < : ICH and international harmonization score, S ＝ : ethics and data compliance score, S > : national and local policy incentives score, S6: standards and guidelines adoption score, S9: risk event and compliance record score.

10. The orthopedic related drug patent value assessment system of claim 1, wherein, 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: The value ranking and grade division unit is configured to rank all the comprehensive value scores of the patents to be evaluated output by the orthopedics-related drug patent comprehensive value evaluation model by percentile, and calculate the percentile ranking of each sample in descending order; The result 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 orthopedics-related drug patent comprehensive value evaluation model; The patent value analysis report generation unit forms a systematic orthopedics-related drug patent value analysis report based on the evaluation results output by the value ranking and grade division unit and / or the orthopedics-related drug patent comprehensive value evaluation model.

Images