Intelligent medical examination equipment and reagent use monitoring and cost early warning management system
By combining the graph mapping module and the dynamic loss model, the problems of broken links and lagging risk monitoring in the operation and management of hospital laboratories have been solved, realizing intelligent cost monitoring and early warning, and improving the real-time performance and accuracy of management.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- CHINA JAPAN FRIENDSHIP HOSPITAL
- Filing Date
- 2025-09-19
- Publication Date
- 2026-06-02
Smart Images
Figure CN121191709B_ABST
Abstract
Description
Technical Field
[0001] This invention belongs to the field of smart medical technology, and in particular relates to a smart medical testing equipment and reagent usage monitoring and cost early warning management system. Background Technology
[0002] Smart healthcare utilizes modern information technology and digitalization to intelligently transform medical processes, services, and management, aiming to improve the quality and efficiency of healthcare services and reduce costs. Currently, the operation and management of hospital laboratories are generally based on a traditional information architecture of "multiple separate systems connected manually." Based on this architecture, hospitals have long faced the following three challenges in achieving lean operations in their laboratories:
[0003] (1) The business-finance link is broken, and a closed loop cannot be formed:
[0004] The data from the five key stages—prescription, testing, billing, cost, and revenue—are fragmented, preventing the formation of a visualized and traceable closed loop in the process of "doctor prescribing—testing execution—medical insurance billing—cost collection—revenue accounting." Management is forced to rely on outdated and coarse-grained financial statements and experience-based judgments, making it difficult to accurately identify loss-making items or high-cost processes, directly impacting the scientific nature of performance evaluation and cost control.
[0005] (2) Project mapping relies on manual work, which is time-consuming and error-prone:
[0006] The naming conventions for chargeable items, testing packages, and reagents and consumables are inconsistent across different systems, with common instances of synonyms being used interchangeably. Current technology can only perform one-by-one matching by manually maintaining mapping tables or rule dictionaries, which is not only extremely labor-intensive but also results in a persistently high rate of mismatches and omissions as new items and consumables are continuously added, seriously affecting subsequent cost accounting and medical insurance compliance.
[0007] (3) Lack of dynamic risk monitoring leads to high costs for post-event remediation:
[0008] The existing system lacks the ability to monitor reagent consumption and medical insurance compliance risks online in real time. Problems such as abnormal reagent consumption and illegal splitting of charges often only come to light during end-of-month inventory or medical insurance audits. By then, the economic losses and compliance risks are already irreversible, hindering the laboratory from further implementing refined management and cost control. With the continuous increase in testing volume and the rapid expansion of data dimensions, the traditional "manual configuration + manual verification" model is no longer sufficient to cope with complex, dynamic, and high-concurrency business scenarios. Summary of the Invention
[0009] To address the problems existing in the prior art, this invention provides an intelligent medical testing equipment and reagent usage monitoring and cost early warning management system, which solves the problem of broken links that prevent the formation of a closed loop, resulting in lagging and inaccurate monitoring.
[0010] This disclosure provides an intelligent medical testing equipment and reagent usage monitoring and cost early warning management system, including:
[0011] The graph mapping module is used to automatically perform semantic parsing and dynamic association of acquired multi-source heterogeneous medical data through a trained language model to obtain a mapping relationship graph.
[0012] The correlation and penetration module is used to build a correlation and penetration analysis framework based on the mapping relationship map, thereby penetrating each medical order income down to the smallest granularity of the test items.
[0013] The cost collection and allocation module is used to collect and allocate the costs of the inspection items based on the dynamic loss model.
[0014] The monitoring module is used to achieve intelligent, transparent monitoring of revenue, costs, and profits based on mapping relationship graphs, correlation and penetration analysis frameworks, and aggregated and allocated costs.
[0015] Optionally, the mapping relationship map is continuously and dynamically optimized, and the mapping relationship map includes:
[0016] A mapping diagram of the relationships between doctors' orders, test items, medical insurance charges, test packages, testing equipment, and testing departments.
[0017] Optionally, the automatic semantic parsing and dynamic association of the acquired multi-source heterogeneous medical data with a trained language model includes:
[0018] Analysis and association of test items, their alternative names, common names, and English abbreviations;
[0019] The medical insurance fee catalog and various packages were broken down and then linked together;
[0020] Many-to-many mapping between reagents / consumables and testing items;
[0021] Analysis and correlation of testing equipment and testing items.
[0022] Optionally, the framework for constructing association and penetration analysis based on the mapping relationship graph, thereby penetrating each medical order revenue step by step down to the smallest granularity of the test items, includes:
[0023] The process involves a step-by-step approach, starting from the laboratory department, equipment, testing packages, and fee items, and finally down to the specific testing procedure.
[0024] Optionally, the dynamic loss model is achieved through a two-layer collaboration between a theoretical benchmark generation engine and a dynamic loss feedback engine.
[0025] The theoretical benchmark generation engine is used to dynamically collect the actual service volume of each test item based on the penetrating analysis framework and generate theoretical consumption benchmark values.
[0026] The dynamic loss feedback engine constructs a dynamic loss feedback function based on real-time acquired actual service volume data.
[0027] Optionally, the formula for the theoretical benchmark generation engine is:
[0028] ,
[0029] in, This is the theoretical benchmark value. The theoretical number of tests per kit provided to reagent manufacturers. As an environmental degradation factor, The loss deviation is caused by differences in operator skill levels, which affects the fluctuation coefficient of the operation. For equipment efficiency index, The time decay factor, This represents the batch variation index.
[0030] Optionally, the dynamic loss feedback function is:
[0031] ,
[0032] in, This is the loss feedback value. To verify the actual number of project reports submitted by the system in real time, For systematic error factor, Instantaneous fluctuations caused by environmental noise.
[0033] Optionally, the dynamic loss model, through a two-layer collaboration between a theoretical benchmark generation engine and a dynamic loss feedback engine, includes:
[0034] When reagent inventory or service volume changes, the theoretical baseline generation engine and the dynamic loss feedback engine automatically trigger calculations to achieve real-time updates of the loss status. Based on historical loss feedback data, correction parameters for theoretical performance parameters are generated. Based on the correction parameters, the error between the baseline value and the actual reagent is reduced, forming a dynamic convergence closed loop.
[0035] Optionally, the cost collection and allocation of the inspection items based on the dynamic loss model includes:
[0036] Indirect costs are integrated with direct material costs, and the integration results, along with the collection and allocation results, form a complete cost view for departments, equipment, and projects.
[0037] Optionally, the intelligent transparent monitoring of revenue, cost, and profit based on mapping relationship graphs, correlation and penetration analysis frameworks, and aggregated and allocated costs includes:
[0038] Build an end-to-end visual view to track the value flow between revenue, costs, and benefits;
[0039] By using time series analysis and machine learning algorithms, key indicators of the system are predicted, and the prediction results are used for budget preparation and forward-looking management.
[0040] The constructed view analyzes the cost composition and departmental contribution, and highlights and warns of abnormal costs.
[0041] The intelligent medical testing equipment and reagent usage monitoring and cost early warning management system provided by this invention obtains a mapping relationship map by mapping heterogeneous medical data, thereby linking various data. Through the correlation and penetration analysis framework, the cost is penetrated to the smallest granularity of the test items, and then the cost is collected and allocated to ensure the authenticity and real-time nature of the cost. Based on the correlation mapping relationship map, each link of medical treatment is detected to achieve the purpose of real-time and accuracy. Attached Figure Description
[0042] The above and other objects, features and advantages of this disclosure will become more apparent from the accompanying drawings, in which like reference numerals generally denote like parts.
[0043] Figure 1 A schematic diagram of the intelligent medical testing equipment and reagent usage monitoring and cost early warning management system provided in this embodiment of the disclosure;
[0044] Figure 2 This is an application block diagram of the intelligent medical testing equipment and reagent usage monitoring and cost early warning management system provided in the embodiments of this disclosure. Detailed Implementation
[0045] The embodiments of this disclosure will now be described in detail with reference to the accompanying drawings.
[0046] It should be understood that the following specific examples illustrate the implementation of this disclosure, and those skilled in the art can easily understand other advantages and effects of this disclosure from the content disclosed in this specification. Obviously, the described embodiments are only a part of the embodiments of this disclosure, and not all of them. This disclosure can also be implemented or applied through other different specific implementation methods, and the details in this specification can also be modified or changed based on different viewpoints and applications without departing from the spirit of this disclosure. It should be noted that, in the absence of conflict, the following embodiments and features in the embodiments can be combined with each other. Based on the embodiments in this disclosure, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this disclosure.
[0047] It should be noted that various aspects of embodiments within the scope of the appended claims are described below. It will be apparent that the aspects described herein can be embodied in a wide variety of forms, and any particular structure and / or function described herein is merely illustrative. Based on this disclosure, those skilled in the art will understand that one aspect described herein can be implemented independently of any other aspect, and two or more of these aspects can be combined in various ways. For example, any number of aspects set forth herein can be used to implement the device and / or practice the method. Additionally, this device and / or method can be implemented using structures and / or functionalities other than one or more of the aspects set forth herein.
[0048] It should also be noted that the illustrations provided in the following embodiments are only schematic representations of the basic concept of this disclosure. The illustrations only show the components related to this disclosure and are not drawn according to the number, shape and size of the components in actual implementation. In actual implementation, the form, quantity and proportion of each component can be arbitrarily changed, and the layout of the components may also be more complex.
[0049] Furthermore, specific details are provided in the following description to facilitate a thorough understanding of the examples. However, those skilled in the art will understand that the described aspects can be practiced without these specific details.
[0050] The technical terms used in this embodiment are explained as follows:
[0051] a) The HIS (Hospital Information System) is responsible for issuing medical orders and settling bills;
[0052] b) The Laboratory Information System (LIS) is only responsible for testing execution and result publication;
[0053] c) The medical insurance settlement subsystem operates independently and completes the cost breakdown according to the medical insurance catalog;
[0054] d) Cost data such as reagents and consumables, human resources, and fixed asset depreciation are scattered across multiple independent modules such as SPD, personnel, and assets.
[0055] The lack of unified data standards and real-time interaction mechanisms among these systems creates typical information silos.
[0056] This embodiment addresses the long-standing systemic bottlenecks in lean operation and intelligent management within the medical testing field. It proposes an intelligent medical testing equipment and reagent usage monitoring and cost early warning management system. Its overall goal is to deeply integrate cutting-edge large-scale modeling technology with medical information systems, completely streamlining the entire business chain from "medical orders to testing to billing to cost to revenue," achieving efficient data aggregation, accurate correlation, and real-time insight. This will provide hospital management with quantifiable, traceable, and predictable decision-making support. Specific objectives are as follows:
[0057] Objective 1: To penetrate data silos and build a comprehensive and precise quantitative system for operational efficiency.
[0058] Currently, hospital systems such as HIS, LIS, medical insurance settlement, and asset management operate independently, resulting in fragmented data. This makes it impossible for managers to scientifically attribute the true profit and loss of individual test items, single pieces of equipment, or even individual departments. This embodiment aims to establish a cross-system data fusion hub, weaving discrete data into a continuous, closed chain of evidence to achieve a precise perspective across all dimensions of "revenue-cost-benefit," providing quantitative support for resource allocation and strategic decision-making.
[0059] Objective 2: To innovate the mapping paradigm and build a semantic-driven automated association engine.
[0060] Traditional solutions rely on manually maintained dictionaries or rule tables to map test items, medical insurance billing items, reagents, consumables, and equipment. This approach is costly, prone to mismatches, and difficult to adapt to frequent changes. This embodiment introduces a large-scale language model deeply optimized for the medical vertical field to build an intelligent mapping engine with medical semantic understanding and logical reasoning capabilities. By replacing "manual rules" with "AI self-learning," it achieves highly accurate, automated, and dynamic associations between multi-source heterogeneous data.
[0061] Objective 3: Reshape risk control concepts and establish a forward-looking, proactive, and intelligent early warning system.
[0062] The existing management model is severely lagging in identifying risks such as abnormal reagent loss and illegal splitting of charges, often only being discovered during post-event audits, resulting in economic losses and compliance risks. This embodiment constructs a data-driven dynamic monitoring and intelligent early warning model to detect key operational indicators in real time 24 / 7, shifting risk identification to the moment business occurs, and realizing a paradigm shift from "post-event remediation" to "pre-event prevention."
[0063] Specifically, such as Figure 1 As shown, this embodiment discloses an intelligent medical testing equipment and reagent usage monitoring and cost early warning management system, including:
[0064] The graph mapping module is used to automatically perform semantic parsing and dynamic association of acquired multi-source heterogeneous medical data through a trained language model to obtain a mapping relationship graph.
[0065] The mapping relationship map is continuously and dynamically optimized, and the mapping relationship map includes:
[0066] A mapping diagram of the relationships between doctors' orders, test items, medical insurance charges, test packages, testing equipment, and testing departments.
[0067] In a specific scenario, such as Figure 2 As shown, a stable, two-way data channel is established with the hospital's HIS, LIS, medical insurance settlement, financial and asset management systems through standard APIs or direct database connection technology, thereby acquiring multi-source heterogeneous medical data; the extracted data is automatically cleaned, formatted, encoded uniformly, deduplicated, and key fields are accurately mapped to build a highly consistent and pure centralized analytical data lake. Figure 2 In the expression n:m, n and m represent different quantities, 1:n represents a one-to-many mapping relationship, and n:m represents a many-to-many mapping relationship. Figure 2 LeaMS in this embodiment refers to the intelligent medical testing equipment and reagent usage monitoring and cost early warning management system disclosed in this embodiment.
[0068] A large language model (LLM) fine-tuned for the medical vertical domain is employed. This model possesses medical semantic understanding and contextual reasoning capabilities, and can automatically identify and associate:
[0069] Test items and their alternative names, common names, and English abbreviations;
[0070] Medical insurance fee schedule and breakdown of various packages;
[0071] Many-to-many mapping between reagents / consumables and testing items;
[0072] The relationship between testing equipment and the execution of testing items.
[0073] Through continuous learning and iterative optimization, we achieve automated, high-coverage, and high-accuracy data association, providing a reliable data foundation for upper-layer applications.
[0074] This implementation utilizes a collaborative mapping architecture combining the semantic understanding capabilities of a publicly disclosed Large Language Model (LLM) with secondary fine-tuning in the medical vertical domain and a self-developed code matching rule engine. This architecture enables multi-dimensional, many-to-many precise semantic alignment between clinical laboratory test items, medical insurance billing items, and in vitro diagnostic reagents. Addressing semantic gaps in medical terminology, such as inconsistent naming, heterogeneous expressions, synonyms (e.g., "complete blood count" and "full blood cell count"), and synonymous names, this implementation introduces the Tongyi LLM. Through its powerful context awareness and medical domain knowledge generalization capabilities, it performs semantic parsing and standardized representation of unstructured or semi-structured item names.
[0075] Based on this, a configurable and traceable "code matching rule engine" is constructed by combining a pre-set industry knowledge rule base, coding standards (such as medical insurance ICDs and medical service item catalogs) and historical code matching experience. This engine performs dynamic collaborative verification and fusion decision-making with the LLM output results, realizing automated matching of the cost collection path from the original test items to medical insurance reimbursable items, and then to the underlying reagents and consumables.
[0076] The correlation and penetration module is used to build a correlation and penetration analysis framework based on the mapping relationship map, thereby penetrating each medical order income down to the smallest granularity of the test items.
[0077] In a specific scenario, refined revenue attribution and full-chain tracing based on the "five-level association system" are employed.
[0078] The five-level correlation and penetrating analysis framework, through the five-level correlation system of "laboratory department → laboratory equipment → laboratory package → fee item → laboratory item", enables the layer-by-layer drilling down and precise attribution of revenue from macro to micro.
[0079] The aforementioned framework for correlation and penetration analysis, based on a mapping relationship graph, penetrates down each medical order revenue step by step to the smallest granularity of the testing items, including:
[0080] The process involves a step-by-step approach, starting from the laboratory department, equipment, testing packages, and fee items, and finally down to the specific testing procedure.
[0081] The cost collection and allocation module is used to collect and allocate the costs of the inspection items based on the dynamic loss model.
[0082] The dynamic loss model is achieved through a two-layer collaboration between a theoretical benchmark generation engine and a dynamic loss feedback engine.
[0083] The theoretical benchmark generation engine is used to dynamically collect the actual service volume of each test item based on the penetrating analysis framework and generate theoretical consumption benchmark values.
[0084] The dynamic loss feedback engine constructs a dynamic loss feedback function based on real-time acquired actual service volume data.
[0085] Optionally, the dynamic loss model, through a two-layer collaboration between a theoretical benchmark generation engine and a dynamic loss feedback engine, includes:
[0086] When reagent inventory or service volume changes, the theoretical baseline generation engine and the dynamic loss feedback engine automatically trigger calculations to achieve real-time updates of the loss status. Based on historical loss feedback data, correction parameters for theoretical performance parameters are generated. Based on the correction parameters, the error between the baseline value and the actual reagent is reduced, forming a dynamic convergence closed loop.
[0087] The cost of the aforementioned inspection items is collected and allocated based on a dynamic loss model, including:
[0088] Indirect costs are integrated with direct material costs, and the integration results, along with the collection and allocation results, form a complete cost view for departments, equipment, and projects.
[0089] Specifically, labor costs (allocated according to staffing and equipment hours), equipment depreciation (allocated according to service life and utilization rate), water and electricity, training and other indirect costs are integrated with direct material costs to form a "complete cost" view for departments, equipment and projects, ensuring that cost analysis is comprehensive and fair.
[0090] This embodiment achieves high-precision allocation of reagent costs and real-time anomaly detection through the dual-layer collaboration of the "theoretical benchmark generation engine" and the "dynamic loss feedback engine".
[0091] The "Theoretical Benchmark Generation Engine" integrates the theoretical performance parameters of reagents (such as the theoretical service volume of a single reagent kit) and dynamically collects the actual service volume of each test item using a multi-level data association architecture to generate theoretical consumption benchmark values.
[0092] ,
[0093] in, This is the theoretical benchmark value. The theoretical number of tests per kit provided to reagent manufacturers. The environmental degradation factor refers to the rate of reagent efficacy degradation caused by fluctuations in environmental temperature and humidity. ); The loss deviation caused by the difference in operator skill level is the fluctuation coefficient. ); The equipment performance index is used to check the aging degree and calibration status of the equipment. state( ） ; The time decay factor refers to the time-dependent decay of the reagent after it has been opened. , (This refers to the time it takes to open the bottle). The batch variation index represents the stability differences between different reagent batches (following an N(0,σ2) distribution).
[0094] By using LLM semantic alignment technology to eliminate the naming heterogeneity of test items (such as "complete blood count" and "full blood cell count"), the statistical homology between theoretical values and actual service volume is ensured.
[0095] The "dynamic loss feedback engine" refers to the real-time access to the actual service volume data of the LIS to construct a dynamic loss feedback function.
[0096] Loss feedback value ,
[0097] in, This is the loss feedback value. To verify the actual number of project reports submitted by the system in real time; The systematic error factor represents the drift in instrument detection accuracy and the rate of false negatives in reports. ; It represents the instantaneous fluctuations (following a Rayleigh distribution) caused by random environmental disturbances, such as electromagnetic interference and vibration.
[0098] The dynamism is reflected in the fact that when reagent inventory changes or service volume increases, the engine automatically triggers incremental calculations to achieve near real-time updates of loss status. Based on historical loss feedback data, the system generates correction suggestions for theoretical performance parameters, gradually bringing the baseline value closer to the actual reagent production capacity, forming a dynamic convergence closed loop.
[0099] The dynamic loss model has achieved three breakthroughs:
[0100] Logical interpretability: The design adopts explicit mathematical relationships (non-fixed formulas) to ensure that the computational logic is traceable.
[0101] Dynamic response capability: Incremental collaborative calculations triggered by changes in reagent inventory status and new service volume events overcome the lag of traditional batch calculation models.
[0102] Self-optimization space: The theoretical performance parameters have the ability to be gradually optimized. Through feedback from historical data, they automatically converge to form correction suggestions that approximate the actual output level, which can avoid static preset deviations.
[0103] Anomaly Warning: Tiered warning mechanism: For example, under a strong intervention strategy, when the loss feedback value deviates significantly from the reasonable range (such as exceeding the threshold T1), the system automatically freezes the associated resource process and pushes the verification task to the management end; under a weak intervention strategy, when the loss feedback value continues to deviate moderately (such as exceeding the threshold T2 for consecutive periods), the system generates a set of optimization suggestions (such as equipment calibration, operation process optimization, etc.).
[0104] The monitoring module is used to achieve intelligent, transparent monitoring of revenue, costs, and profits based on mapping relationship graphs, correlation and penetration analysis frameworks, and aggregated and allocated costs.
[0105] Based on mapping relationship graphs, correlation and penetration analysis frameworks, and aggregated and allocated costs, intelligent transparent monitoring of revenue, costs, and profits is achieved, including:
[0106] Construct a visual view that tracks the value flow between revenue, cost, and benefit from end-to-end; in a specific embodiment, a Sankey diagram can be constructed to track the value flow from end-to-end "revenue → cost → benefit".
[0107] By employing time series analysis and machine learning algorithms, key indicators of the system are predicted, and the prediction results are used for budget preparation and forward-looking management. Trend prediction and comparison with the same period can provide decision support for budget preparation and forward-looking management.
[0108] The constructed view analyzes cost composition and departmental contribution, and highlights and alerts abnormal costs. Stacked area charts and pie charts can be used to analyze cost composition and departmental contribution, and highlight and alert on abnormal costs.
[0109] This embodiment provides a panoramic view of revenue, allowing managers to drill down with one click across any dimension (department, equipment, package, individual item) to view the revenue composition in real time; the system can adaptively degrade analysis in scenarios with missing data to ensure business continuity.
[0110] Provides an integrated, interactive analytics dashboard in a specific scenario, including three main dashboards: revenue, cost, and profit, supporting unlimited drill-down levels from the entire hospital down to departments, equipment, packages, and individual items.
[0111] This embodiment has the following advantages:
[0112] 1. The system eliminates information silos between laboratory testing fees, reagent costs, and medical insurance settlements. By connecting multi-dimensional data links between testing items, medical insurance fees, reagents, consumables, and equipment, it comprehensively presents the true relationship between testing items, actual costs, medical insurance settlements, and revenue. Laboratory departments no longer need to rely on experience to guess which items are profitable and which are unprofitable; the system can clearly calculate the "income and expenditure" of individual testing items, giving departments a clear understanding and enabling more scientific operations.
[0113] 2. AI-powered automatic mapping eliminates tedious processes and reduces errors. It eliminates the need for manual maintenance of massive project dictionaries and manual matching of names and pinyin to determine charges or medical insurance items. The large language model, like a knowledgeable laboratory technician, understands the true meaning of different names, combinations, and packages, automatically completing multi-dimensional mapping from test items to medical insurance charges to costs. This saves significant manpower and avoids mismatches and omissions that easily occur with manual matching.
[0114] 3. The system proactively alerts users to any anomalies detected, mitigating risks in advance. In the past, many reagent leaks and unreasonable itemized billing issues only surfaced during internal audits. This system can monitor theoretical usage and actual consumption in real time, automatically comparing historical data with reasonable ranges to quickly identify abnormal reagent losses and potential issues like duplicate billing, providing early warnings and truly managing risks before they occur and preventing them at their source.
[0115] 4. Comprehensive closed-loop analysis helps the laboratory department to be more confident in operational audits and performance evaluations. The system can visualize the entire process from doctor's prescription, test execution, medical insurance billing to final revenue, and intuitively display income, costs, and profits using multi-dimensional dashboards, Sankey diagrams, cost stacking analysis, and other formats. No matter which project, equipment, or department the leader asks about the effectiveness, the laboratory department can clearly provide charts and graphs with sound reasoning.
[0116] 5. Achieve lean management throughout the entire equipment lifecycle to maximize asset benefits. The system elevates equipment management to a strategic level. It not only accurately calculates the revenue, cost, and profit of individual pieces of equipment, clearly demonstrating their "economic value," but also deeply assesses their "operational efficiency" by analyzing utilization rates and dynamic wear and tear, promptly identifying issues such as aging and calibration. Combined with AI-powered fault prediction and risk warning, it achieves a shift from "passive maintenance" to "predictive maintenance," ultimately providing data-driven decision support throughout the entire equipment lifecycle for procurement, operation, allocation, and even retirement, ensuring that every investment in assets generates maximum benefit.
[0117] 6. Flexible adaptation to big data scenarios involving multiple hospital campuses and departments: Traditional rule-based maintenance methods are almost unsustainable when dealing with multiple hospital campuses and a large number of different testing and medical insurance catalogs. Based on a large language model that is deeply optimized for medical scenarios, it can learn and adapt as the data scale expands, easily handling the complex needs of multiple hospital campuses, multiple standards, and multiple versions of medical insurance catalogs. The more the system is used, the "smarter" it becomes.
[0118] 7. Ultimate Goal: Cost reduction, efficiency improvement, and compliance. By meticulously monitoring every consumption and charge of testing reagents, we can help the laboratory plug the "invisible loopholes" and improve the compliance of medical insurance cost control. This will enable the laboratory to not only stand firm in its scientific and technical disciplines but also demonstrate its value in business management and performance evaluation.
[0119] This embodiment intelligently constructs a mapping relationship between medical insurance billing items and laboratory tests, reagents, and consumables using a Large Language Model (LLM) that has undergone secondary optimization in the medical field. This achieves high-accuracy automatic matching, replacing manual comparison methods. A "five-level association system" is proposed for revenue and cost analysis, establishing a complete link between doctor's orders, laboratory tests, medical insurance billing, equipment, and departments. A closed-loop, visualized analysis system is constructed between medical insurance billing, actual costs, and laboratory test revenue, supporting hospitals in conducting refined performance evaluations and operational optimization. A dynamic loss rate calculation model is introduced to achieve intelligent comparison between theoretical and actual usage and reasonable cost allocation. Individual costs are dynamically adjusted based on the actual loss rate of each item, thus affecting subsequent revenue analysis and performance ranking. This enables automatic warnings of abnormal loss rates that lower project revenue, effectively monitoring resource waste. A degradation operation mechanism is supported, automatically switching analysis dimensions when data is incomplete, ensuring the continuity and stability of system analysis.
[0120] The system in this embodiment can be integrated with the DRG / DIP payment model and can be further extended to interface with the DRG (Diagnosis Related Groups) or DIP (Disease-Specific Value Payment) system to realize cost-benefit assessment and cost control strategy optimization by disease, and enhance medical insurance compliance analysis capabilities.
[0121] The system in this embodiment supports both cloud deployment and local deployment modes, adapting to different hospital information strategies and data security needs, and has broad applicability.
[0122] The system in this embodiment introduces machine learning algorithms to predict future revenue trends, cost trends, and equipment failure probabilities for different projects based on historical data, thereby achieving forward-looking operation management.
[0123] The system in this embodiment can be expanded to manage multiple campuses and centers in a unified manner, supporting centralized analysis and horizontal comparison at the group level, and helping to improve the efficiency of regional resource allocation.
[0124] The basic principles of this disclosure have been described above with reference to specific embodiments. However, it should be noted that the advantages, benefits, and effects mentioned in this disclosure are merely examples and not limitations, and should not be considered as essential features of each embodiment of this disclosure. Furthermore, the specific details disclosed above are for illustrative and facilitative purposes only, and are not limitations. These details do not limit the scope of this disclosure to the necessity of employing the aforementioned specific details for implementation.
[0125] In this disclosure, relational terms such as "first" and "second" are used merely to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. The block diagrams of devices, apparatuses, devices, and systems involved in this disclosure are merely illustrative examples and are not intended to require or imply that they must be connected, arranged, or configured in the manner shown in the block diagrams. As those skilled in the art will recognize, these devices, apparatuses, devices, and systems can be connected, arranged, and configured in any manner. Words such as "comprising," "including," "having," etc., are open-ended terms meaning "including but not limited to," and are used interchangeably with them. The terms "or" and "and" as used herein refer to the terms "and / or," and are used interchangeably with them unless the context clearly indicates otherwise. The term "such as" as used herein refers to the phrase "such as but not limited to," and is used interchangeably with it.
[0126] Additionally, as used herein, the "or" used in a list of items beginning with "at least one" indicates a separate list, such that a list of, for example, "at least one of A, B, or C" means A or B or C, or AB or AC or BC, or ABC (i.e., A and B and C). Furthermore, the word "exemplary" does not imply that the described example is preferred or better than other examples.
[0127] It should also be noted that in the systems and methods of this disclosure, the components or steps can be decomposed and / or recombined. These decompositions and / or recombinations should be considered as equivalent solutions to this disclosure.
[0128] Various changes, substitutions, and modifications can be made to the technology described herein without departing from the teachings defined by the appended claims. Furthermore, the scope of the claims of this disclosure is not limited to the specific aspects of the processes, machines, manufactures, events, means, methods, and actions described above. Currently existing or later-developed processes, machines, manufactures, events, means, methods, or actions that perform substantially the same function or achieve substantially the same result as the corresponding aspects described herein can be utilized. Therefore, the appended claims include such processes, machines, manufactures, events, means, methods, or actions within their scope.
[0129] The above description of the disclosed aspects is provided to enable any person skilled in the art to make or use this disclosure. Various modifications to these aspects will be readily apparent to those skilled in the art, and the general principles defined herein may be applied to other aspects without departing from the scope of this disclosure. Therefore, this disclosure is not intended to be limited to the aspects shown herein, but rather to be carried out within the widest scope consistent with the principles and novel features disclosed herein.
[0130] The above description has been given for purposes of illustration and description. Furthermore, this description is not intended to limit the embodiments of this disclosure to the forms disclosed herein. Although numerous exemplary aspects and embodiments have been discussed above, those skilled in the art will recognize certain variations, modifications, alterations, additions, and sub-combinations therein.
Claims
1. A smart medical testing equipment and reagent usage monitoring and cost early warning management system, characterized in that, include: The graph mapping module is used to automatically perform semantic parsing and dynamic association of acquired multi-source heterogeneous medical data through a trained language model to obtain a mapping relationship graph. The correlation and penetration module is used to build a correlation and penetration analysis framework based on the mapping relationship map, thereby penetrating each medical order income down to the smallest granularity of the test items. The cost collection and allocation module is used to collect and allocate the costs of the inspection items based on the dynamic loss model. The monitoring module is used to achieve intelligent, transparent monitoring of revenue, costs, and profits based on mapping relationship graphs, correlation and penetration analysis frameworks, and aggregated and allocated costs. The dynamic loss model is achieved through a two-layer collaboration between a theoretical benchmark generation engine and a dynamic loss feedback engine. The theoretical benchmark generation engine is used to dynamically collect the actual service volume of each test item based on the penetrating analysis framework and generate theoretical consumption benchmark values. The dynamic loss feedback engine constructs a dynamic loss feedback function based on real-time acquired actual service volume data. The formula for the theoretical benchmark generation engine is: ， in, This is the theoretical benchmark value. The theoretical number of tests per kit provided to reagent manufacturers. As an environmental degradation factor, The loss deviation is caused by differences in operator skill levels, which affects the fluctuation coefficient of the operation. For equipment efficiency index, The time decay factor, This refers to the batch variation index; The dynamic loss feedback function is: ， in, This is the loss feedback value. To verify the actual number of project reports submitted by the system in real time, For systematic error factor, Instantaneous fluctuations caused by environmental noise.
2. The intelligent medical testing equipment and reagent usage monitoring and cost early warning management system according to claim 1, characterized in that, The mapping relationship map is continuously and dynamically optimized, and the mapping relationship map includes: A mapping diagram of the relationships between doctors' orders, test items, medical insurance charges, test packages, testing equipment, and testing departments.
3. The intelligent medical testing equipment and reagent usage monitoring and cost early warning management system according to claim 1, characterized in that, The automatic semantic parsing and dynamic association of the acquired multi-source heterogeneous medical data with a trained language model includes: Analysis and association of test items, their alternative names, common names, and English abbreviations; The medical insurance fee catalog and various packages were broken down and then linked together; Many-to-many mapping between reagents / consumables and testing items; Analysis and correlation of testing equipment and testing items.
4. The intelligent medical testing equipment and reagent usage monitoring and cost early warning management system according to claim 1, characterized in that, The aforementioned framework for correlation and penetration analysis, based on a mapping relationship graph, penetrates down each medical order revenue step by step to the smallest granularity of the testing items, including: The process involves a step-by-step approach, starting from the laboratory department, equipment, testing packages, and fee items, and finally down to the specific testing procedure.
5. The intelligent medical testing equipment and reagent usage monitoring and cost early warning management system according to claim 1, characterized in that, The dynamic loss model employs a two-layer collaboration between a theoretical benchmark generation engine and a dynamic loss feedback engine, including: When reagent inventory or service volume changes, the theoretical baseline generation engine and the dynamic loss feedback engine automatically trigger calculations to achieve real-time updates of the loss status. Based on historical loss feedback data, correction parameters for theoretical performance parameters are generated. Based on the correction parameters, the error between the baseline value and the actual reagent is reduced, forming a dynamic convergence closed loop.
6. The intelligent medical testing equipment and reagent usage monitoring and cost early warning management system according to claim 1, characterized in that, The cost collection and allocation of the inspection items based on the dynamic loss model includes: Indirect costs are integrated with direct material costs, and the integration results, along with the collection and allocation results, form a complete cost view for departments, equipment, and projects.
7. The intelligent medical testing equipment and reagent usage monitoring and cost early warning management system according to claim 1, characterized in that, The intelligent, transparent monitoring of revenue, costs, and profits, based on mapping relationship graphs, correlation and penetration analysis frameworks, and aggregated and allocated costs, includes: Build an end-to-end visual view to track the value flow between revenue, costs, and benefits; By using time series analysis and machine learning algorithms, key indicators of the system are predicted, and the prediction results are used for budget preparation and forward-looking management. The constructed view analyzes the cost composition and departmental contribution, and highlights and warns of abnormal costs.