Medical equipment intelligent management system and method based on general name and asset unit double-track mapping

Abstract

The invention discloses a medical equipment intelligent management system and method based on general name and asset unit double-track mapping. The system is in butt joint with a UDI database API of the State Drug Administration through a general name mapping engine, and equipment name standardization is achieved; aggregating financial, clinical and operation multi-dimensional data through an asset unit constructor, and constructing an equipment digital twinborn body; establishing a bidirectional mapping relation between the standard general name and the asset unit through a double-track core engine; and generating an optimization decision through the intelligent decision center based on an efficiency comparison algorithm. The method solves the problems of term confusion, data island, decision lag and the like in medical equipment management. Compared with the prior art, the data association speed is improved by 72000 times, the first-aid response time is shortened by 93%, the purchase repetition rate is reduced by 96%, and the review work efficiency is improved by 97%.

Description

Technical Field

[0001] This invention relates to the field of intelligent medical technology, specifically to an intelligent management system and method for medical devices based on a dual-track mapping of generic names and asset units. Background Technology

[0002] Existing medical device management systems face numerous technical challenges in practical applications. The primary issue lies in the lack of standardization in device naming systems, resulting in significant inconsistencies in device names across different information systems. Hospital information systems, financial management systems, and IoT monitoring platforms often use different naming conventions for the same device, leading to difficulties in cross-system data association and the formation of information silos. Statistics show that over 18% of device queries fail due to name mismatches.

[0003] Traditional equipment management schemes have significant shortcomings in benefit assessment. Clinical value data, such as positive detection rate and equipment utilization rate, are disconnected from economic benefit data, such as procurement cost and operating expenses, failing to form a unified comprehensive equipment evaluation system. This makes it difficult for managers to accurately assess the true return on investment of equipment, affecting the scientific rigor of subsequent procurement and configuration decisions.

[0004] Meanwhile, existing technologies have limited capabilities in supporting equipment decision-making. Most medical institutions rely heavily on the subjective experience and judgment of management personnel for equipment procurement, allocation, and disposal decisions, lacking objective decision-making basis based on data analysis. Compliance management efficiency in medical quality assessment is low, with the error rate of manual record-keeping generally exceeding 18%, severely impacting the accuracy and efficiency of the assessment process. Existing technological solutions mostly employ a single-identity management model, failing to establish a multi-dimensional correlation system for equipment and thus unable to meet the actual needs of modern medical institutions for refined equipment management. Summary of the Invention

[0005] The technical problem to be solved by the present invention is to provide a medical device intelligent management system based on a dual-track mapping of generic names and asset units, so as to solve the problems of inconsistent device naming, serious data silos, and insufficient decision support in the prior art.

[0006] To solve the above-mentioned technical problems, the technical solution adopted by the present invention is as follows: A medical device intelligent management system based on a dual-track mapping of generic names and asset units. The system adopts a dual-track parallel technical architecture and achieves intelligent management of medical devices by combining standardized device naming with multi-dimensional data aggregation.

[0007] The system includes a generic name mapping engine module, which is responsible for establishing the mapping relationship between device product names and standard generic names set by the National Medical Products Administration (NMPA). It obtains authoritative standard device name information by connecting to the NMPA's Unique Device Identifier (UDI) database API. This module employs an intelligent matching algorithm based on semantic similarity, which can automatically identify device product names and map them to standard generic names. The mapping algorithm comprehensively considers semantic similarity, edit distance similarity, and contextual relevance, and calculates the optimal matching result through weighted calculation. Specifically, the similarity calculation formula is: Similarity(A,B) = α·Semantic(A,B) + β·Edit(A,B) + γ·Context(A,B) Where α, β, and γ are weighting coefficients, and satisfy... a+b+c=1 Semantic similarity is calculated based on a pre-trained BERT model, edit distance similarity is quantified using the Levenshtein algorithm, and context similarity is calculated using the TF-IDF vector space model.

[0008] This module further employs a dynamic weight adjustment mechanism, optimizing weight allocation based on historical matching success rates. The weight adjustment formula is: α(t+1) = α(t) × (1 + ρ × ΔAccuracy_semantic) β(t+1) = β(t) × (1 + ρ × ΔAccuracy_edit) γ(t+1) = γ(t) × (1 + ρ × ΔAccuracy_context) Where t represents the iteration round, α(t), β(t), and γ(t) represent the weight coefficients of semantic similarity, edit distance, and context in round t, respectively; α(t+1), β(t+1), and γ(t+1) represent the weight coefficients after the update in round t+1; ρ is the learning rate parameter (ranging from 0.01 to 0.1); and ΔAccuracy_semantic, ΔAccuracy_edit, and ΔAccuracy_context represent the changes in accuracy in the semantic similarity, edit distance similarity, and context similarity dimensions, respectively. Through a continuous learning mechanism, the system can continuously optimize matching accuracy and adapt to the evolving trends in medical device naming.

[0009] The system also includes an asset unit builder module, which enables dynamic aggregation and processing of multi-source heterogeneous data. By establishing a dynamic ETL data pipeline, it synchronizes equipment-related data from the hospital information system, financial management system, and IoT sensors in real time. This module constructs digital twins of the equipment, abstracting each device into a data entity containing three core dimensions: procurement cost, real-time status, and clinical value. The procurement cost dimension includes economic indicators such as the initial purchase price, installation and commissioning costs, and training costs; the real-time status dimension collects dynamic information such as equipment operating parameters, fault status, and usage frequency through IoT sensors; and the clinical value dimension statistically analyzes medical quality indicators such as the equipment's positive examination rate, diagnostic accuracy, and patient satisfaction. The asset unit model uses a multi-dimensional vector representation to map the equipment's entire lifecycle information into structured data objects.

[0010] This module employs an asset valuation algorithm to quantitatively assess the overall value of the equipment. The asset valuation formula is as follows: AssetValue(t) = InitialCost × e^(-λt) + Σ[ClinicalBenefit(i) × UtilizationRate(i)] - MaintenanceCost(t) Where AssetValue(t) represents the asset value of the equipment at time t (unit: RMB 10,000), InitialCost represents the initial purchase cost of the equipment (unit: RMB 10,000), λ represents the equipment depreciation rate coefficient, t represents the equipment usage time (unit: years), i represents the sequence number of the i-th clinical use record, ClinicalBenefit(i) represents the clinical benefit value generated by the i-th use, UtilizationRate(i) represents the equipment utilization rate in the i-th time period, and MaintenanceCost(t) represents the cumulative maintenance cost up to time t (unit: RMB 10,000). This algorithm can accurately assess the actual value of the equipment at different time points, providing a scientific basis for equipment upgrades and replacements.

[0011] The asset unit builder also integrates an equipment lifecycle prediction model, using a Weibull distribution function to predict equipment reliability trends. The equipment reliability function is: R(t) = exp[-(t / n)^β] Where t represents the equipment operating time (in hours), R(t) represents the reliability probability value of the equipment at time t, η is the characteristic life parameter (representing the operating time when the equipment reaches the failure probability), and β is the shape parameter (β<1 indicates early failure, β=1 indicates random failure, and β>1 indicates wear failure). By combining historical fault data and operating status information, the system can predict the remaining service life of the equipment, identify equipment requiring maintenance or replacement in advance, and achieve preventative maintenance management.

[0012] The system further includes a dual-track core engine module, which is the core component of the system and is responsible for integrating generic name mapping data and asset unit data to establish a dual identification system for devices. This engine achieves unified retrieval and correlation analysis of cross-system data by constructing an association index between a standard terminology library and a full lifecycle data lake. The dual-track mapping mechanism allows each device to simultaneously possess a standardized generic name identifier and a personalized asset unit identifier, ensuring both the standardization of device information and the preservation of individual device characteristics.

[0013] The system also includes an intelligent decision-making center module, which provides scientific decision support for managers based on a performance comparison algorithm for equipment of the same type. This module employs a multi-dimensional performance evaluation algorithm to calculate the comprehensive performance index of the equipment. The performance index calculation formula is as follows: Performance(i) = Σ(Wi × Ni) The economic dimension has a weight of 30%, calculated as Wi = ROI / 0.3 × 0.3; the clinical dimension has a weight of 40%, calculated as Wi = Positive Rate / 0.6 × 0.4; and the operational dimension has a weight of 30%, calculated as Wi = 1 / (Failure Rate + ε) × 0.3, where ε is a minimal constant to prevent the denominator from being zero. Through performance comparison analysis, the system can identify high-performing and low-performing equipment, providing quantitative basis for procurement decisions, equipment allocation, and replacement / renewal.

[0014] The intelligent decision-making center further integrates equipment configuration optimization algorithms, employing a genetic algorithm to solve for the optimal equipment resource configuration scheme. The objective function for equipment configuration optimization is: Maximize Z = Σ[Pi × Ui × Ei] - Σ[Ci × Xi] The constraints are: Σ[Xi] ≤ Budget, Σ[Si × Xi] ≤ Space, where i represents the sequence number of the i-th equipment type, Z represents the overall system benefit value, Pi represents the efficiency coefficient of the i-th equipment, Ui represents the expected utilization rate of the i-th equipment, Ei represents the single clinical benefit value of the i-th equipment, Ci represents the unit cost of the i-th equipment, Xi represents the number of i-th equipment, Budget represents the budget constraint, Si represents the space occupied by the i-th equipment, and Space represents the available space constraint. This algorithm can find the optimal equipment configuration combination under limited budget and space constraints, maximizing overall medical benefits.

[0015] This module also includes a device fault early warning algorithm, which predicts the probability of device faults based on historical fault data and real-time monitoring parameters using a support vector machine model. The fault early warning function is as follows: P(failure) = σ(Σ[αi × yi × K(xi, x)] + b) Where P(failure) represents the predicted probability of equipment failure (ranging from 0 to 1), and σ is the sigmoid activation function. (σ(z)=1 / (1+e^(-z))) Let ...

[0016] The system ultimately includes an application layer service module, which provides diverse management functions to end users. This includes a procurement feasibility report generation submodule, capable of automatically generating a feasibility analysis report for equipment procurement based on historical data and performance analysis of similar equipment; a real-time dispatch instruction submodule, which provides optimal route planning for equipment allocation by analyzing equipment location, status, and clinical needs; and an automatic review log submodule, which automatically extracts relevant equipment data and generates a review log that meets regulatory requirements based on preset review standards and rules.

[0017] A smart management method for medical devices based on a dual-track mapping of generic names and asset units is proposed. This method achieves intelligent management of medical devices through a systematic data processing flow.

[0018] The first step involves equipment information collection and standardization. By scanning the QR code on the equipment nameplate or directly entering basic equipment information, the system obtains the equipment's product name, model specifications, manufacturer, and other identifying information. Then, it calls the API interface of the National Medical Products Administration's Unique Device Identifier Database to query the corresponding standard generic name. If a direct match cannot be found, an AI-based synonym matching algorithm is activated, using semantic analysis technology to automatically convert the product name to the generic name. This process employs a confidence assessment mechanism; when the matching confidence level is lower than a preset threshold, the system marks it as an item awaiting manual review to ensure the accuracy of the mapping results.

[0019] Next, the asset unit construction phase is executed. A dynamic ETL pipeline for multi-source data is established to achieve real-time synchronization with heterogeneous data sources such as the hospital information system, financial management system, and IoT monitoring platform. Preprocessing steps such as data cleaning, format conversion, and quality verification ensure data consistency and integrity. Based on this, a digital twin of the equipment is constructed, aggregating equipment-related information scattered across various systems into a unified data object. The digital twin includes static attribute information of the equipment, such as purchase time, technical parameters, and maintenance contracts, as well as dynamic status information, such as real-time operating parameters, fault records, and usage statistics.

[0020] Then, the dual-track mapping and association phase is executed. A two-way mapping relationship is established between the generic name identifier and the asset unit identifier, forming a composite identifier system for the equipment. By constructing a distributed index structure, fast retrieval based on the generic name or asset unit ID is supported. Simultaneously, cross-system data association rules are established to achieve automatic matching and synchronous updating of equipment information in different business systems. This mapping relationship supports dynamic maintenance; when equipment information changes, the system automatically updates the relevant mapping records.

[0021] Finally, the intelligent decision generation stage begins. Based on the completed dual-track mapping data, a performance comparison analysis of equipment within the same category is initiated. The system categorizes and aggregates equipment according to its standard generic name, calculating the performance distribution and statistical characteristics of similar equipment. Through horizontal comparative analysis, individual equipment with abnormal performance is identified, and the causes of these abnormalities are analyzed. Based on the analysis results, the system automatically generates corresponding management suggestions, including equipment procurement recommendations, allocation optimization plans, maintenance reminders, and phase-out warnings. These suggestions are presented to managers through a visual interface, supporting an interactive decision-making process.

[0022] The beneficial effects of this invention are reflected in the following aspects: Regarding data association efficiency, the dual-track mapping mechanism significantly improves the performance of cross-system data retrieval, reducing data association speed from the traditional 4 hours to 0.2 seconds, an efficiency improvement of 72,000 times. Regarding emergency response capabilities, the intelligent scheduling algorithm can quickly locate the optimal equipment and plan delivery routes, reducing emergency equipment dispatch time from the traditional 43 minutes to 3 minutes, a response speed improvement of 93%. Regarding the accuracy of procurement decisions, the decision support mechanism based on performance comparison of similar equipment reduces the equipment duplication rate from 8.7% to 0.3%, improving decision accuracy by 96%. Regarding compliance management efficiency, the automated review ledger generation function reduces the human error rate from 18% to 0.5%, improving compliance work efficiency by 97%. Attached Figure Description

[0023] Figure 1 This is a flowchart illustrating the overall system architecture of the present invention; Figure 2 This is a flowchart of the generic name mapping engine of the present invention; Figure 3 This is a flowchart illustrating the asset unit construction process of the present invention. Detailed Implementation

[0024] To make the objectives, technical solutions, and advantages of this invention clearer, the following embodiments provide a more detailed description of the invention. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of the invention.

[0025] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be described in detail below with reference to specific embodiments.

[0026] A large tertiary hospital has deployed the intelligent medical equipment management system of this invention to achieve unified management of more than 2,800 medical devices throughout the hospital.

[0027] like Figure 1 As shown in the diagram, the overall architecture flowchart of the system of this invention illustrates the complete processing architecture from data acquisition to decision output. The top layer of the system is the data source layer, which includes multi-source data inputs such as data sources, HIS / LIS / PACS systems, financial systems, and IoT sensors. These data are first aggregated into a unified data lake for centralized storage. Then, they enter the dual-track core engine module for split processing: the left branch uses the generic name mapping engine module to convert the equipment product name into a standard generic name and store it in the standard terminology library; the right branch uses the asset unit builder module to aggregate multi-dimensional data to build a digital twin of the equipment and store it in the full lifecycle data lake. The data from the two tracks converge in the intelligent decision center module for performance comparison analysis and decision generation. Finally, the application layer service module provides users with three core functions: procurement demonstration reports, real-time scheduling instructions, and automatic review ledgers. The diagram clearly shows the technical architecture and data flow of the dual-track parallel processing.

[0028] like Figure 2 As shown in the flowchart, the generic name mapping engine details the processing steps for device name standardization. The process begins with device information input; the system first checks if a cached mapping exists. If it does, the mapping result is returned directly; otherwise, the system calls the National Medical Products Administration's UDI database API interface to query the standard name. Then, semantic similarity calculation and a synonym AI matching algorithm are initiated, achieving optimal matching by calculating a weighted score of semantic similarity, edit distance similarity, and contextual similarity. Next, a confidence assessment is performed to determine if the confidence level exceeds a preset threshold. When the confidence level exceeds the threshold, the mapping cache is automatically updated; when the confidence level is below the threshold, it is marked as a manually reviewed item. In either case, the system ultimately updates the mapping cache and returns the mapping result, outputting the standard generic name. The diagram clearly marks each processing node, decision branch, and data flow.

[0029] like Figure 3As shown in the flowchart, the asset unit construction process illustrates the creation of a digital twin of equipment. The process begins with multi-source data acquisition; then, data standardization is achieved through data cleaning and preprocessing; next, the data flow is divided into three parallel processing branches: the left side is financial dimension analysis, including procurement cost calculation and operating cost analysis; the middle side is clinical dimension analysis, including positive rate statistics and equipment utilization rate calculation; and the right side is operation and maintenance dimension analysis, including failure rate analysis and maintenance cost statistics. The results of these three dimensions are converged into the asset unit model, forming a digital twin of the equipment. Subsequently, an efficiency index is calculated to comprehensively evaluate equipment performance based on multi-dimensional data; finally, decision recommendations are generated to provide quantitative decision support for managers. The diagram clearly illustrates the complete data processing flow from multi-source heterogeneous data to three-dimensional parallel analysis, and then to efficiency evaluation and decision generation.

[0030] The hospital's original equipment management system relied on traditional manual ledger recording, with equipment information scattered across multiple platforms including the Hospital Information System (HIS), Laboratory Information System (LIS), Picture Archiving and Communication System (PACS), financial system, and equipment department management system. This resulted in severe information inconsistencies and difficulties in retrieval. By implementing the technical solution of this invention, the hospital established a unified equipment management platform based on dual-track mapping.

[0031] In the initial stage of system deployment, the standardization of equipment information was carried out first. Technicians used mobile terminal devices to scan the QR codes on the nameplates of existing hospital equipment one by one to collect basic identification information. For a GE Discovery CT750 HD computed tomography scanner, the system first recorded its product name as "DiscoveryCT750 HD". Then, by calling the UDI database API interface of the National Medical Products Administration, the system found that the standard generic name corresponding to the equipment was "X-ray computed tomography equipment". The system's generic name mapping engine used a semantic similarity algorithm to calculate the matching degree between the product name and the standard name. In this case, the semantic similarity was 0.92, the edit distance similarity was 0.78, the context similarity was 0.85, and the comprehensive similarity score reached 0.87, exceeding the system's set threshold of 0.8. Therefore, the mapping relationship was automatically established.

[0032] During the asset unit construction phase, the system collected relevant data for the CT equipment from multiple source systems via an ETL data pipeline. The system extracted the equipment's procurement cost information from the financial system, including a purchase price of 4.6 million yuan, installation and commissioning costs of 150,000 yuan, and personnel training costs of 80,000 yuan, calculating a total procurement cost of 4.83 million yuan. Clinical usage data for the equipment was collected from the HIS system, including an annual examination volume of 3,240 patients, a positive detection rate of 78.5%, and an average examination time of 25 minutes. Equipment operating status data was obtained from the IoT monitoring system, including an operating time of 8,760 hours, 12 failures, and an average interval between failures of 730 hours. Based on this multi-dimensional data, the system constructed an asset unit model for the equipment, forming a unique asset unit identifier "CT_GE_750HD".

[0033] The dual-track core engine establishes a mapping relationship between the equipment's generic name and asset unit identifier, i.e., "X-ray computed tomography equipment" ←→ "CT_GE_750HD". Through this dual-track mapping mechanism, administrators can retrieve all similar equipment using the standard generic name, and also query detailed information for a specific equipment using the asset unit identifier. The system establishes a distributed index structure, supporting millisecond-level cross-system data retrieval.

[0034] The intelligent decision-making center module performed a performance evaluation analysis on the CT equipment. In terms of economics, the annual return on investment was calculated to be 18.5%, ranking 3rd among similar equipment. In terms of clinical applications, its positive detection rate of 78.5% exceeded the industry average of 72%, demonstrating excellent performance. In terms of operation and maintenance, the failure rate was 1.37%, lower than the average failure rate of 2.1% for similar equipment, indicating good control of operation and maintenance costs. Based on the multi-dimensional performance evaluation algorithm, the equipment's comprehensive performance index was 85.7 points, ranking 1st among similar equipment in the entire hospital, and the system marked it as a high-performance device.

[0035] In practical applications, when an emergency department needs to schedule an urgent CT scan for a patient suspected of having a stroke, the traditional approach requires medical staff to contact the radiology department by phone to confirm equipment availability, followed by manual patient transfer arrangements, a process that typically takes 35-45 minutes. With this invention, emergency physicians submit examination requests via a terminal device. The system's intelligent decision center immediately queries all available equipment under the "X-ray Computed Tomography Equipment" category. Through real-time status monitoring, it determines that the CT_GE_750HD equipment is currently idle, closest to the emergency department, and has optimal performance. The system automatically generates the optimal route plan and sends dispatch instructions to relevant departments. From request submission to patient arrival at the examination room, the entire process is reduced to 8 minutes, significantly improving emergency response efficiency.

[0036] In terms of supporting equipment procurement decisions, when a hospital plans to add a new CT scanner, the system automatically generates a performance comparison report of similar equipment. By analyzing the historical operating data of the existing three CT scanners, the system found that the average annual number of examinations per scanner is 3,150, the average return on investment is 16.8%, and the average failure rate is 1.85%. Based on this data analysis, the system recommends that the technical parameters of the new equipment should not be lower than the configuration level of the current best-in-class equipment, the budget should be controlled between 4.5 and 5 million yuan, and the expected annual number of examinations should reach more than 3,000 to ensure the return on investment. This data-driven procurement recommendation significantly improves the scientific nature and accuracy of equipment procurement decisions.

[0037] In terms of compliance management, the hospital's quality management department is required to submit a medical equipment management review report to the National Health Commission every quarter. Traditionally, staff had to manually extract data from multiple systems and compile ledgers, a process that was extremely labor-intensive and prone to errors. With the system of this invention, the generation of review ledgers is fully automated. Based on preset review standards, the system automatically extracts key indicator data for each piece of equipment, including equipment availability, utilization rate, maintenance timeliness, and safety incident records. For this CT equipment, the automatically generated ledger shows: equipment availability 100%, annual utilization rate 88.7%, preventative maintenance execution rate 100%, and zero safety incident records. The entire ledger generation process has been reduced from three working days to 30 minutes, significantly improving work efficiency.

[0038] As can be seen from the detailed description of this embodiment, the present invention effectively solves the key technical problems in medical equipment management through the dual-track mapping technology architecture, realizes the standardization and unification of equipment information, the intelligent aggregation of multi-dimensional data, and the automated support for scientific decision-making, bringing significant efficiency improvement and quality enhancement to the equipment management work of medical institutions.

Claims

1. A medical device intelligent management system based on a dual-track mapping of generic names and asset units, characterized in that, The system includes a generic name mapping engine module, which interfaces with the National Medical Products Administration's Medical Device Unique Identifier database API to convert device product names into standard generic names. It employs a semantic similarity-based intelligent matching algorithm, using weighted calculations of semantic similarity, edit distance similarity, and contextual relevance to achieve optimal matching. An asset unit builder module is used to establish a dynamic ETL data pipeline, aggregating multi-source data from hospital information systems, financial systems, and IoT sensors in real time to construct a digital twin of the equipment, encompassing three dimensions: procurement cost, real-time status, and clinical value. An intelligent decision-making center module, based on a performance comparison algorithm for equipment of the same category, uses a multi-dimensional weighted calculation method to generate optimized decision instructions for equipment procurement, allocation, and disposal.

2. The intelligent management system for medical devices based on a dual-track mapping of generic names and asset units according to claim 1, characterized in that, The semantic similarity algorithm in the generic name mapping engine module adopts the following calculation formula: Similarity(A,B) = α·Semantic(A,B) + β·Edit(A,B) + γ·Context(A,B), where Similarity(A,B) represents the comprehensive similarity score between the device product name A and the standard generic name B, A represents the device product name, B represents the standard generic name, α, β, and γ are weight coefficients and satisfy α+β+γ=1, Semantic(A,B) represents the semantic similarity calculated based on the pre-trained BERT model, Edit(A,B) represents the edit distance similarity calculated using the Levenshtein algorithm, and Context(A,B) represents the contextual similarity calculated using the TF-IDF vector space model. When the comprehensive similarity exceeds a preset threshold, a mapping relationship is automatically established; when it is below the threshold, it is marked as an item awaiting manual review.

3. The intelligent management system for medical devices based on a dual-track mapping of generic names and asset units according to claim 1, characterized in that, The asset unit builder module maps the entire lifecycle information of the equipment into structured data objects using multidimensional vector representation. The procurement cost dimension includes economic indicators such as the initial purchase price of the equipment, installation and commissioning costs, and training costs. The real-time status dimension collects dynamic information such as equipment operating parameters, fault status, and usage frequency through IoT sensors. The clinical value dimension statistically analyzes the equipment's positive examination rate, diagnostic accuracy, and patient satisfaction medical quality indicators to form a unique asset unit identifier for the equipment.

4. The intelligent management system for medical devices based on a dual-track mapping of generic names and asset units according to claim 1, characterized in that, It also includes a dual-track core engine module, which is used to integrate generic name mapping data and asset unit data to establish a dual identification system for devices. By building a standard terminology library and a correlation index of the full lifecycle data lake, it enables unified retrieval and correlation analysis of cross-system data. The dual-track mapping mechanism enables each device to have both a standardized generic name identifier and a personalized asset unit identifier, supports fast retrieval based on generic name or asset unit ID, and supports dynamic maintenance and automatic updates of the mapping relationship.

5. The intelligent management system for medical devices based on a dual-track mapping of generic names and asset units according to claim 1, characterized in that, The intelligent decision-making center module uses a multi-dimensional performance evaluation algorithm to calculate the comprehensive performance index of equipment. The calculation formula is Performance(i) = Σ(Wi × Ni), where Performance(i) represents the comprehensive performance index of the i-th equipment, Wi represents the weight value of the i-th dimension, and Ni represents the normalized score of the i-th dimension. The economic dimension has a weight of 30% and the calculation formula is Wi = ROI / 0.3 × 0.

3. The clinical dimension has a weight of 40% and the calculation formula is Wi = positive rate / 0.6 × 0.

4. The operation and maintenance dimension has a weight of 30% and the calculation formula is Wi = 1 / (failure rate + ε) × 0.3, where ε is a minimal constant to prevent the denominator from being zero. Through performance comparison analysis, excellent and poor equipment are identified, providing a quantitative basis for procurement decisions, equipment allocation, and replacement.

6. The intelligent management system for medical devices based on a dual-track mapping of generic names and asset units according to claim 1, characterized in that, It also includes an application layer service module, which includes a procurement feasibility report generation submodule that automatically generates equipment procurement feasibility analysis reports based on historical data and performance analysis of similar equipment. The real-time scheduling instruction submodule provides optimal path planning for equipment allocation by analyzing equipment location, status, and clinical needs; the automatic review log submodule automatically extracts relevant equipment data and generates a review log that meets regulatory requirements based on preset review standards and rules, thereby automating the review process.

7. A method for intelligent management of medical devices based on a dual-track mapping of generic names and asset units, characterized in that, This includes the equipment information collection and standardization process. By scanning the QR code on the equipment nameplate, the product name identification information of the equipment is obtained. The corresponding standard generic name is queried by calling the API interface of the National Medical Products Administration's Medical Device Unique Identifier Database. When a direct match cannot be found, an AI-based synonym matching algorithm is activated to automatically convert the product name to the generic name. During the asset unit construction phase, a dynamic ETL pipeline for multi-source data is established to achieve real-time synchronization of heterogeneous data sources. A digital twin of the equipment is constructed through data cleaning, format conversion, and quality verification preprocessing steps.

8. The intelligent management method for medical devices based on a dual-track mapping of generic names and asset units according to claim 7, characterized in that, It also includes a dual-track mapping and association stage, which establishes a two-way mapping relationship between generic name identifiers and asset unit identifiers to form a composite identifier system for equipment. By constructing a distributed index structure, it supports fast retrieval functions based on generic names or asset unit IDs. It establishes cross-system data association rules to realize automatic matching and synchronous updates of equipment information in different business systems. When equipment information changes, the system automatically updates the relevant mapping records to ensure the consistency and accuracy of the mapping relationship.

9. The intelligent management method for medical devices based on a dual-track mapping of generic names and asset units according to claim 7, characterized in that, It also includes an intelligent decision generation stage, which initiates a performance comparison analysis of equipment of the same category based on dual-track mapping data. The system classifies and aggregates equipment according to its standard generic name to calculate the performance distribution and statistical characteristics of similar equipment. Through horizontal comparative analysis, it identifies individual equipment with abnormal performance and analyzes the reasons for the abnormality. Based on the analysis results, it automatically generates equipment procurement recommendations, allocation optimization plans, maintenance reminders, and elimination early warning management suggestions, which are presented to managers through a visual interface to support the interactive decision-making process.

10. The intelligent management method for medical devices based on a dual-track mapping of generic names and asset units according to claim 7, characterized in that, The synonym matching algorithm adopts a confidence evaluation mechanism. When the matching confidence is lower than a preset threshold, the system marks it as an item to be manually reviewed to ensure the accuracy of the mapping results. The digital twin contains static attribute information and dynamic status information of the equipment. The static attribute information includes purchase time, technical parameters, and maintenance contracts. The dynamic status information includes real-time operating parameters, fault records, and usage statistics. Through multi-dimensional data aggregation, unified management and intelligent analysis of the equipment's entire life cycle information are realized.

Images