A DRG-based medical consumable use evaluation method and system

Abstract

The application discloses a medical consumable use evaluation method and system based on DRG, and belongs to the technical field of medical data analysis, and comprises the following steps: obtaining medical consumable use data of historical patients in a target DRG group; constructing consumable use behavior graphs of each medical execution unit based on the medical consumable use data; taking each medical execution unit as an evaluation object, performing comprehensive scoring based on a multi-dimensional evaluation vector of a cost dimension, a quality dimension and an efficiency dimension, selecting a medical execution unit with a comprehensive score meeting a preset condition to construct a benchmark unit set; taking a medical execution unit not selected into the benchmark unit set as a unit to be optimized, matching a benchmark unit with the highest patient portrait similarity to each unit to be optimized as a learning benchmark; comparing and analyzing the consumable use behavior graphs of the unit to be optimized and the corresponding learning benchmark, identifying key consumable categories to be optimized, and generating an evaluation result; and the method can improve the accuracy of consumable use evaluation.

Description

Technical Field

[0001] This application relates to the field of medical data analysis technology, and in particular to a method and system for evaluating the use of medical consumables based on DRG. Background Technology

[0002] Diagnosis Related Groups (DRGs) are a method of grouping hospitalized patients based on factors such as diagnosis, surgery, age, and comorbidities. They are widely used in medical insurance payments and healthcare quality assessment. Under the DRG payment model, cost control and rational use of medical consumables become crucial for healthcare institution management.

[0003] Existing methods for evaluating the use of consumables mostly rely on total statistics or single-indicator comparisons, lacking systematic modeling of consumable usage behavior. This makes it difficult to identify the root causes of differences between different medical execution units (such as departments or doctor groups) and also fails to provide benchmarks and specific directions for improvement for units to be optimized. Summary of the Invention

[0004] To address the aforementioned issues, this application provides a method and system for evaluating the use of medical consumables based on DRG, aiming to solve the technical problem that existing technologies, which involve differences among patients in different medical execution units, directly comparing the quantity or amount of consumables used in each unit to evaluate the use of medical consumables, can affect the accuracy of the evaluation.

[0005] This application provides a method for evaluating the use of medical consumables based on DRG, including: Obtain historical medical consumables usage data for patients within the target DRG group; Based on the medical consumables usage data, a consumables usage behavior map of each medical execution unit is constructed; Taking each medical execution unit as the evaluation object, a comprehensive score is calculated based on a multi-dimensional evaluation vector of cost, quality, and efficiency dimensions. Medical execution units that meet the preset conditions in terms of comprehensive score are selected to construct a set of benchmark units. Medical execution units that are not selected into the benchmark unit set are designated as units to be optimized, and each unit to be optimized is matched with the benchmark unit with the highest patient profile similarity as a learning benchmark. The consumable usage behavior graphs of the unit to be optimized and the corresponding learning benchmarks are compared and analyzed to identify the key consumable categories to be optimized and generate evaluation results.

[0006] In some embodiments, constructing a consumable usage behavior map for each medical execution unit based on the medical consumable usage data includes: Medical consumables are categorized into several consumable categories, with each consumable category serving as a node. Each node is assigned basic attributes, including the quantity and cost of the consumables used in that category. For each patient, directed edges are constructed according to the order of use, and bidirectional edges are constructed according to co-occurrence relationships; Aggregate all patient-level nodes and edges belonging to the same medical execution unit: take statistical values ​​for the basic attributes of nodes, use the transfer probability of directed edges as weights according to direction, and use the co-occurrence frequency of bidirectional edges as weights to obtain the consumable usage behavior map of the unit.

[0007] In some embodiments, classifying medical consumables into several consumable categories includes: Medical consumables should be classified according to the medical insurance classification code first. For medical consumables not covered by the medical insurance classification code, they should be classified according to their clinical functional attributes.

[0008] In some embodiments, the calculation steps of the comprehensive score include: The multidimensional evaluation vector is normalized. The objective weights are obtained by using the entropy weight method, the subjective weights are obtained by using the analytic hierarchy process, and the objective weights and subjective weights are weighted and averaged to obtain the combined weights of each dimension. The normalized evaluation vector is weighted and summed according to the combined weights to obtain the comprehensive score of each medical execution unit.

[0009] In some embodiments, the selection of medical execution units that meet preset criteria for comprehensive scores to construct a benchmark unit set includes: The medical execution units are ranked from highest to lowest according to the comprehensive score. The benchmark unit set consists of medical execution units whose overall score is greater than a preset threshold and whose scores in each dimension are not lower than the average score of all medical execution units in that dimension.

[0010] In some embodiments, when the number of medical execution units whose comprehensive score is greater than a preset threshold and whose scores in each dimension are not lower than the average score of all medical execution units in that dimension is zero, the top N medical execution units in terms of comprehensive score are selected to form the benchmark unit set.

[0011] In some embodiments, the method for constructing a patient profile includes: Extract numerical and categorical features from the patient's medical record cover page; For each numerical feature, the membership vector is calculated using a predefined fuzzy set and membership function; for each categorical feature, the membership degree of each code is calculated according to the mapping rule between categories and fuzzy sets, and the maximum value is taken on each fuzzy set to obtain the membership vector. The membership vectors of all features are concatenated to form a fuzzy profile vector for the patient. Then, the fuzzy profile vectors of all patients within the same medical execution unit are averaged to obtain the patient profile for that unit.

[0012] In some embodiments, the step of comparing and analyzing the consumable usage behavior maps of the unit to be optimized with those of the corresponding learning benchmark to identify key consumable categories to be optimized and generating evaluation results includes: Node matching is performed between the graph of the unit to be optimized and the graph of the learning benchmark to determine the common node set; For each node in the common node set, calculate the difference value of the node's basic attributes as the node's deviation degree, and calculate the difference value of the weight of the edge that the node participates in as the edge deviation penalty value. The node deviation is obtained by weighted summing of the node deviation and the edge deviation penalty value. The consumable categories corresponding to nodes with a comprehensive deviation greater than a preset threshold are the key consumable categories to be optimized. Evaluation results are generated based on the aforementioned key consumable categories.

[0013] In some embodiments, the medical execution unit is a single clinical department, a group of surgeons, or a treatment team.

[0014] This invention provides a DRG-based medical consumables usage evaluation system, comprising: The acquisition module is used to acquire historical medical consumables usage data of patients within the target DRG group; The module is used to construct a consumable usage behavior map for each medical execution unit based on the medical consumable usage data. The scoring module is used to evaluate each medical execution unit based on a multi-dimensional evaluation vector of cost, quality and efficiency dimensions, and select medical execution units whose comprehensive scores meet preset conditions to build a set of benchmark units. The matching module is used to identify medical execution units that are not included in the set of benchmark units as units to be optimized, and to match each unit to be optimized with the benchmark unit with the highest patient profile similarity as a learning benchmark. The analysis module is used to compare and analyze the consumable usage behavior map of the unit to be optimized with the corresponding learning benchmark, identify the key consumable categories to be optimized, and generate evaluation results.

[0015] Compared with the prior art, the present invention has the following beneficial effects: By eliminating confounding factors through patient profile matching, expressing consumable usage patterns through structured behavioral mapping, scientifically selecting benchmarks through multi-dimensional weighting, and outputting accurate evaluation results through mapping comparison analysis, the accuracy of consumable usage evaluation can be improved. Attached Figure Description

[0016] The embodiments of the present invention will be further described below with reference to the accompanying drawings: Figure 1 A schematic diagram illustrating the implementation process of a DRG-based method for evaluating the use of medical consumables, provided in an embodiment of this application. Figure 2 A schematic diagram of the structure of a DRG-based medical consumables usage evaluation system provided in this application embodiment; Figure 3 This is a schematic diagram of the composition structure of the electronic device provided in the embodiments of this application. Detailed Implementation

[0017] To make the objectives, technical solutions, and advantages of this application clearer, the application will be further described in detail below with reference to the accompanying drawings. The described embodiments should not be regarded as limitations on this application. All other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0018] In the following description, references are made to “some embodiments,” which describe a subset of all possible embodiments. However, it is understood that “some embodiments” may be the same subset or different subsets of all possible embodiments and may be combined with each other without conflict.

[0019] If the application documents contain similar descriptions such as "first, second, third", the following explanation shall be added: In the following description, the terms "first, second, third" are used only to distinguish similar objects and do not represent a specific order of objects. It is understood that "first, second, third" may be interchanged in a specific order or sequence where permitted, so that the embodiments of this application described herein can be implemented in an order other than that illustrated or described herein.

[0020] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used herein is for the purpose of describing embodiments of this application only and is not intended to limit this application.

[0021] To address the problems existing in related technologies, this application provides a method for evaluating the use of medical consumables based on DRG. The subject executing this evaluation method can be an electronic device. The electronic device can be various types of terminals such as laptops, tablets, desktop computers, set-top boxes, and mobile devices (e.g., mobile phones, portable music players, personal digital assistants, dedicated messaging devices, portable gaming devices), or it can be implemented as a server. The server can be a standalone physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server providing basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, content delivery networks (CDNs), and big data and artificial intelligence platforms.

[0022] In some embodiments, the functions implemented by the evaluation method provided in this application can be achieved by the processor of an electronic device calling program code, wherein the program code can be stored in a computer storage medium.

[0023] This application provides a method for evaluating the use of medical consumables based on DRG. Figure 1 This application provides a schematic diagram of the implementation process of a DRG-based method for evaluating the use of medical consumables, as illustrated in the embodiments below. Figure 1 As shown, it includes: Step S1: Obtain historical medical consumables usage data for patients within the target DRG group; In this embodiment of the invention, a specific DRG group to be analyzed is first selected, such as a DRG group related to a cardiac interventional treatment. From the hospital information system (HIS) and consumables management system, the hospitalization records and consumable usage details of all discharged patients in this group over a past period are extracted, including fields such as consumable name, specifications, medical insurance code, quantity used, unit price, and usage time. By limiting the analysis to the same DRG group, the confounding effects of patient disease complexity and treatment differences on consumable usage evaluation are eliminated, laying the foundation for subsequent horizontal comparability between units.

[0024] Step S2: Based on the medical consumables usage data, construct a consumables usage behavior map for each medical execution unit; In some embodiments, step S2 includes: Step S21: Classify medical consumables into several consumable categories, and assign basic attributes to each consumable category as a node. The basic attributes include the quantity and cost of use for that consumable category. Step S22: For each patient, construct directed edges according to the usage sequence and bidirectional edges according to the co-occurrence relationship; Step S23: Aggregate all patient-level nodes and edges belonging to the same medical execution unit: take statistical values ​​for the basic attributes of the nodes, calculate the transfer probability of directed edges according to direction as weight, and calculate the co-occurrence frequency of bidirectional edges as weight to obtain the consumable usage behavior map of the unit.

[0025] In this embodiment of the invention, medical consumables are categorized into several categories, with each category serving as a node. Basic attributes are assigned to each node, including the quantity and cost of use for that category. Categorization prioritizes the use of medical insurance classification codes. For consumables not covered by these codes, clinical functional attributes are used. This ensures compatibility with national medical insurance standards while covering all consumables actually used in clinical practice. For each patient, directed edges are constructed according to usage sequence (e.g., the category of consumables used previously points to the category used later), and bidirectional edges are constructed based on co-occurrence relationships (common use within the same hospitalization cycle). This captures the sequential logic and combination patterns of consumable usage. All patient-level nodes and edges belonging to the same medical execution unit (e.g., a single clinical department, surgical team, or treatment team) are aggregated: statistical values ​​(e.g., mean, median, or mode) are taken for the basic attributes of the nodes; the direction-based transition probability is used as a weight for directed edges; and the co-occurrence frequency is used as a weight for bidirectional edges, thus obtaining the consumable usage behavior graph for that unit. Behavioral mapping transforms consumable usage from isolated statistical indicators into structured behavioral patterns, which can intuitively show the preferences and patterns of each medical execution unit in the selection, order, and combination of consumables.

[0026] In this embodiment of the invention, taking a specific clinical department as an example, for all target DRG patients admitted to that department, the usage sequence of consumables for each patient is transformed into directed edges (e.g., "mapping catheter - ablation catheter"), and bidirectional edges are established for consumable pairs used in the same surgery. During aggregation, the average number and cost of "mapping catheters" used by patients in that department are calculated, the proportion of patients who switched from "mapping catheters" to "ablation catheters" is used as the weight of the directed edges, and the frequency of patients who co-occurred "mapping catheters" and "catheter sheaths" is used as the weight of the bidirectional edges. Finally, a graph is formed with consumable categories as nodes, weighted directed edges, and weighted bidirectional edges.

[0027] Step S3: Using each medical execution unit as the evaluation object, a comprehensive score is calculated based on a multi-dimensional evaluation vector encompassing cost, quality, and efficiency dimensions. Medical execution units whose comprehensive scores meet preset conditions are selected to construct a set of benchmark units. In some embodiments, the calculation steps of the comprehensive score include: Step S31: Normalize the multidimensional evaluation vector; Step S32: Obtain objective weights using the entropy weight method, obtain subjective weights using the analytic hierarchy process, and then perform a weighted average of the objective and subjective weights to obtain the combined weights of each dimension. Step S33: The normalized evaluation vector is weighted and summed according to the combined weights to obtain the comprehensive score of each medical execution unit.

[0028] In some embodiments, the selection of medical execution units that meet preset criteria for comprehensive scores to construct a benchmark unit set includes: Step S34: Sort each medical execution unit from high to low according to the comprehensive score; Step S35: Select medical execution units whose comprehensive score is greater than the preset threshold and whose scores in each dimension are not lower than the average score of all medical execution units in that dimension, and form the benchmark unit set.

[0029] In this embodiment of the invention, the multi-dimensional evaluation vector consisting of cost, quality, and efficiency dimensions is normalized to eliminate the dimensional differences between indicators of different dimensions and ensure the comparability of evaluation results across dimensions. Normalization can employ common methods such as min-max normalization and z-score normalization, with the appropriate method selected based on the characteristics of the indicators. The objective weights for each dimension are obtained using the entropy weight method, and the subjective weights are obtained using the analytic hierarchy process (AHP). The objective and subjective weights are then weighted and averaged to obtain the combined weights for each dimension. During the weighted averaging process, the proportions of objective and subjective weights can be adjusted according to actual needs to ensure a scientific and reasonable weight allocation. Based on the determined combined weights, the normalized multi-dimensional evaluation vector is weighted and summed to obtain the comprehensive score for each medical execution unit. A higher comprehensive score indicates a better overall level of consumable usage for that medical execution unit. The medical execution units are ranked from highest to lowest comprehensive score to initially select units with superior overall performance. A benchmark unit set is formed by selecting medical execution units whose overall score is greater than a preset threshold and whose scores in each dimension are not lower than the average score of all medical execution units in that dimension. The preset threshold can be flexibly set according to the management objectives and industry standards of the medical institution; the requirement that the scores in each dimension are not lower than the average score ensures that the benchmark units have no obvious shortcomings in the three dimensions of cost, quality, and efficiency, and have comprehensive reference value.

[0030] In this embodiment of the invention, it is assumed that there are multiple clinical departments as medical execution units. The average cost of consumables per visit (cost), complication rate (quality), and average length of hospital stay (efficiency) are calculated for each department, and normalized to obtain scores for each dimension. Data-driven objective weights are calculated using the entropy weight method, and subjective weights are obtained through expert scoring. These weighted weights are then combined to obtain a comprehensive score for each department. A threshold is set, and it is checked whether the scores for each dimension are not lower than the average score of all departments in that dimension. Departments that meet the criteria are selected into the benchmark unit set.

[0031] In some embodiments, when the number of medical execution units whose comprehensive score is greater than a preset threshold and whose scores in each dimension are not lower than the average score of all medical execution units in that dimension is zero, the top N medical execution units in terms of comprehensive score are selected to form the benchmark unit set.

[0032] In this embodiment of the invention, to avoid the number of medical execution units in the benchmark unit set being zero, the units ranked in the top N of the comprehensive scores can be selected as the candidate benchmark set, where N is a natural number greater than zero.

[0033] Step S4: The medical execution units that are not selected into the benchmark unit set are taken as units to be optimized, and the benchmark unit with the highest patient profile similarity is matched as the learning benchmark for each unit to be optimized; In some embodiments, the method for constructing a patient profile includes: Step S41: Extract numerical features and categorical features from the patient's medical record cover page; Step S42: For each numerical feature, calculate the membership vector using a predefined fuzzy set and membership function; for each categorical feature, calculate the membership degree of each code according to the mapping rule between categories and fuzzy sets, and take the maximum value on each fuzzy set to obtain the membership vector. Step S43: Concatenate the membership vectors of all features into a fuzzy profile vector for the patient, and then average the fuzzy profile vectors of all patients within the same medical execution unit to obtain the patient profile for that unit.

[0034] In this embodiment of the invention, the patient profile is represented using fuzzy sets. Each patient's features belong to multiple fuzzy sets, and the membership vector is used as the profile. The specific construction method is as follows: Numerical features (age, comorbidity index CCI score) and categorical features (diagnosis code set, surgical operation code set) are extracted from the information on the first page of the patient's medical record. For each numerical feature, several fuzzy sets and their membership functions are predefined (e.g., the fuzzy set for age can be defined as "young", "middle-aged", and "old", and the membership function can be a triangular membership function). The membership degree of the feature value under each membership function is calculated to obtain the membership vector of the feature. For each categorical feature, according to the mapping rules between categories and fuzzy sets (e.g., the mapping between diagnosis codes and disease severity fuzzy sets), the membership degree of each category in the feature set to each fuzzy set is determined. Then, the maximum value of the membership degree of all categories is taken on each fuzzy set to obtain the membership vector of the feature. The membership vectors of all features are concatenated to obtain the fuzzy profile vector of the patient. For each medical execution unit, the fuzzy profile vectors of all patients treated by the unit are averaged and aggregated to obtain the patient profile of the medical execution unit. Then, cosine similarity is used to calculate the similarity between the patient profiles of the unit to be optimized and each benchmark unit, and the benchmark unit with the highest similarity is selected as the learning benchmark for the unit to be optimized. Through fuzzy matching of patient profiles, the differences in consumable usage patterns caused by differences in patient groups (such as age and severity of disease) are effectively eliminated, making the comparison between the unit to be optimized and the learning benchmark fair.

[0035] In this embodiment of the invention, it is assumed that department A was not selected into the benchmark set and belongs to the unit to be optimized. Patients in department A have a higher average age and a heavier burden of comorbidities, while the age distribution and comorbidity distribution of patients in benchmark unit C are highly similar to those in department A. By calculating the cosine similarity using fuzzy profile vectors, it is found that department A and department C have the highest similarity; therefore, department C is used as the learning benchmark for department A. Conversely, if the patient characteristics of department A differ significantly from those of department B, department B will not be mistakenly used as the benchmark.

[0036] Step S5: Compare and analyze the consumable usage behavior map of the unit to be optimized with that of the corresponding learning benchmark to identify the key consumable categories to be optimized and generate evaluation results.

[0037] In some embodiments, step S5 includes: Step S51: Perform node matching between the graph of the unit to be optimized and the graph of the learning benchmark to determine the common node set; Step S52: For each node in the common node set, calculate the difference value of the node's basic attributes as the node deviation, and calculate the weight difference value of the edges that the node participates in as the edge deviation penalty value. Step S53: The node deviation degree and the edge deviation penalty value are weighted and summed to obtain the comprehensive deviation degree of the node; Step S54: The consumable category corresponding to the node with a comprehensive deviation greater than the preset threshold is the key consumable category to be optimized; Step S55: Generate evaluation results based on the key consumable categories.

[0038] In this embodiment of the invention, the graph of the unit to be optimized is matched with the graph of the learning benchmark to determine a common node set. For each node in the common node set, the difference in the node's basic attributes (average usage quantity per person, average usage amount per person) is calculated as the node deviation degree; simultaneously, the weight difference between the directed edges and bidirectional edges in which the node participates is calculated as the edge deviation penalty value (for example, when the transition probability of a directed edge decreases beyond a threshold, a penalty is added to the two endpoint nodes of that edge). The node deviation degree and the edge deviation penalty value are weighted and summed to obtain the comprehensive deviation degree of the node. The consumable categories corresponding to nodes with a comprehensive deviation degree greater than a preset threshold are identified as key consumable categories to be optimized. Based on the key consumable categories and their deviation reasons (such as high usage quantity, high usage amount, failure to use necessary consumables, excessive use of non-centralized procurement brands, etc.), specific evaluation results and improvement suggestions are generated. Graph comparison analysis not only points out which consumable categories have problems, but also reveals the specific manifestations of the problems (quantity problem or amount problem) through node deviation degree, and reveals differences in behavioral patterns (such as unreasonable timing or improper combination) through edge deviation penalty value. This fine-grained diagnostic capability is something that traditional total quantity comparison methods cannot provide. The final evaluation results are clearly actionable and can directly guide clinical departments to improve their consumable usage practices.

[0039] In this embodiment of the invention, the consumables behavior graph of department A (to be optimized) is aligned with the graph of benchmark department C. The average usage quantity and cost per person for each consumable category are compared. It is found that department A's average usage quantity and cost for "ablation catheters" are significantly higher than department C's. Simultaneously, comparing directed edge weights reveals that department A's transition probability from "mapping catheters" to "ablation catheters" is similar to department C's, but department A uses a large number of "imported brand ablation catheters" not used by department C. After considering node deviation and edge deviation penalties, the overall deviation of "ablation catheters" exceeds the threshold and is identified as a key consumable to be optimized. The evaluation result output is: "Excessive use of ablation catheters, a tendency to choose high-priced imported brands; it is recommended to refer to the benchmark department and prioritize the use of centrally procured brands, and establish a single-case, single-catheter usage standard." Based on the foregoing embodiments, this application provides a DRG-based medical consumables usage evaluation system. The system includes various modules and units, which can be implemented by a processor in a computer device; of course, it can also be implemented by specific logic circuits. In the implementation process, the processor can be a central processing unit (CPU), a microprocessor (MPU), a digital signal processor (DSP), or a field programmable gate array (FPGA), etc.

[0040] This application provides a DRG-based medical consumables usage evaluation system. Figure 2 A schematic diagram of a DRG-based medical consumables usage evaluation system provided in this application embodiment is shown below. Figure 2 As shown, it includes: The acquisition module is used to acquire historical medical consumables usage data of patients within the target DRG group; The module is used to construct a consumable usage behavior map for each medical execution unit based on the medical consumable usage data. The scoring module is used to evaluate each medical execution unit based on a multi-dimensional evaluation vector of cost, quality and efficiency dimensions, and select medical execution units whose comprehensive scores meet preset conditions to build a set of benchmark units. The matching module is used to identify medical execution units that are not included in the set of benchmark units as units to be optimized, and to match each unit to be optimized with the benchmark unit with the highest patient profile similarity as a learning benchmark. The analysis module is used to compare and analyze the consumable usage behavior map of the unit to be optimized with the corresponding learning benchmark, identify the key consumable categories to be optimized, and generate evaluation results.

[0041] It should be noted that, in the embodiments of this application, if the above evaluation method is implemented as a software functional module and sold or used as an independent product, it can also be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the embodiments of this application, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), magnetic disks, or optical disks. Thus, the embodiments of this application are not limited to any specific hardware and software combination.

[0042] Accordingly, this application provides a storage medium storing a computer program thereon, characterized in that the computer program, when executed by a processor, implements the steps in the evaluation method provided in the above embodiments.

[0043] This application provides an electronic device; Figure 3 This is a schematic diagram of the composition structure of the electronic device provided in the embodiments of this application, such as... Figure 3 As shown, the electronic device 100 includes: a processor 101, at least one communication bus 102, a user interface 103, at least one external communication interface 104, and a memory 105. The communication bus 102 is configured to enable communication between these components. The user interface 103 may include a display screen, and the external communication interface 104 may include standard wired and wireless interfaces. The processor 101 is configured to execute a program of an evaluation method stored in the memory to implement the steps of the evaluation method provided in the above embodiment.

[0044] It should be noted that the descriptions of the storage media and electronic device embodiments above are similar to the descriptions of the method embodiments above, and have similar beneficial effects. For technical details not disclosed in the storage media and device embodiments of this application, please refer to the descriptions of the method embodiments of this application for understanding.

[0045] It should be understood that the phrase "one embodiment" or "an embodiment" throughout the specification means that a specific feature, structure, or characteristic related to the embodiment is included in at least one embodiment of this application. Therefore, "in one embodiment" or "in an embodiment" appearing throughout the specification does not necessarily refer to the same embodiment. Furthermore, these specific features, structures, or characteristics can be combined in any suitable manner in one or more embodiments. It should be understood that in the various embodiments of this application, the sequence numbers of the above-described processes do not imply a sequential order of execution; the execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this application. The sequence numbers of the above-described embodiments are merely descriptive and do not represent the superiority or inferiority of the embodiments.

[0046] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, object, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, object, or apparatus. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, object, or apparatus that includes that element.

[0047] In the several embodiments provided in this application, it should be understood that the disclosed devices and methods can be implemented in other ways. The device embodiments described above are merely illustrative. For example, the division of units is only a logical functional division, and in actual implementation, there may be other division methods, such as: multiple units or components can be combined, or integrated into another system, or some features can be ignored or not executed. In addition, the coupling, direct coupling, or communication connection between the various components shown or discussed can be through some interfaces, and the indirect coupling or communication connection between devices or units can be electrical, mechanical, or other forms.

[0048] The units described above as separate components may or may not be physically separate. The components shown as units may or may not be physical units. They may be located in one place or distributed across multiple network units. Some or all of the units may be selected to achieve the purpose of this embodiment according to actual needs.

[0049] In addition, each functional unit in the various embodiments of this application can be integrated into one processing unit, or each unit can be a separate unit, or two or more units can be integrated into one unit; the integrated unit can be implemented in hardware or in the form of hardware plus software functional units.

[0050] Those skilled in the art will understand that all or part of the steps of the above method embodiments can be implemented by hardware related to program instructions. The aforementioned program can be stored in a computer-readable storage medium. When the program is executed, it performs the steps of the above method embodiments. The aforementioned storage medium includes various media that can store program code, such as mobile storage devices, read-only memory (ROM), magnetic disks, or optical disks.

[0051] Alternatively, if the integrated units described above are implemented as software functional modules and sold or used as independent products, they can also be stored in a computer-readable storage medium. Based on this understanding, the technical solutions of the embodiments of this application, or the parts that contribute to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a controller to execute all or part of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as mobile storage devices, ROMs, magnetic disks, or optical disks.

[0052] The above description is merely an embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A method for evaluating the use of medical consumables based on DRG, characterized in that, include: Obtain historical medical consumables usage data for patients within the target DRG group; Based on the medical consumables usage data, a consumables usage behavior map of each medical execution unit is constructed; Taking each medical execution unit as the evaluation object, a comprehensive score is calculated based on a multi-dimensional evaluation vector of cost, quality, and efficiency dimensions. Medical execution units that meet the preset conditions in terms of comprehensive score are selected to construct a set of benchmark units. Medical execution units that are not selected into the benchmark unit set are designated as units to be optimized, and each unit to be optimized is matched with the benchmark unit with the highest patient profile similarity as a learning benchmark. The consumable usage behavior graphs of the unit to be optimized and the corresponding learning benchmarks are compared and analyzed to identify the key consumable categories to be optimized and generate evaluation results.

2. The method according to claim 1, characterized in that, The process of constructing a consumable usage behavior map for each medical execution unit based on the medical consumable usage data includes: Medical consumables are categorized into several consumable categories, with each consumable category serving as a node. Each node is assigned basic attributes, including the quantity and cost of the consumables used in that category. For each patient, directed edges are constructed according to the order of use, and bidirectional edges are constructed according to co-occurrence relationships; Aggregate all patient-level nodes and edges belonging to the same medical execution unit: take statistical values ​​for the basic attributes of nodes, use the transfer probability of directed edges as weights according to direction, and use the co-occurrence frequency of bidirectional edges as weights to obtain the consumable usage behavior map of the unit.

3. The method according to claim 2, characterized in that, The classification of medical consumables into several categories includes: Medical consumables should be classified according to the medical insurance classification code first. For medical consumables not covered by the medical insurance classification code, they should be classified according to their clinical functional attributes.

4. The method according to claim 1, characterized in that, The calculation steps for the comprehensive score include: The multidimensional evaluation vector is normalized. The objective weights are obtained by using the entropy weight method, the subjective weights are obtained by using the analytic hierarchy process, and the objective weights and subjective weights are weighted and averaged to obtain the combined weights of each dimension. The normalized evaluation vector is weighted and summed according to the combined weights to obtain the comprehensive score of each medical execution unit.

5. The method according to claim 1, characterized in that, The selection of medical execution units that meet preset criteria for comprehensive scores constitutes a benchmark unit set, including: The medical execution units are ranked from highest to lowest according to the comprehensive score. The benchmark unit set consists of medical execution units whose overall score is greater than a preset threshold and whose scores in each dimension are not lower than the average score of all medical execution units in that dimension.

6. The method according to claim 5, characterized in that, When the number of medical execution units whose comprehensive score is greater than the preset threshold and whose scores in each dimension are not lower than the average score of all medical execution units in that dimension is zero, the top N medical execution units in terms of comprehensive score are selected to form the benchmark unit set.

7. The method according to claim 1, characterized in that, Methods for constructing patient profiles include: Extract numerical and categorical features from the patient's medical record cover page; For each numerical feature, the membership vector is calculated using a predefined fuzzy set and membership function; for each categorical feature, the membership degree of each code is calculated according to the mapping rule between categories and fuzzy sets, and the maximum value is taken on each fuzzy set to obtain the membership vector. The membership vectors of all features are concatenated to form a fuzzy profile vector for the patient. Then, the fuzzy profile vectors of all patients within the same medical execution unit are averaged to obtain the patient profile for that unit.

8. The method according to claim 2, characterized in that, The step involves comparing and analyzing the consumable usage behavior maps of the unit to be optimized with those of the corresponding learning benchmarks to identify key consumable categories to be optimized and generating evaluation results, including: Node matching is performed between the graph of the unit to be optimized and the graph of the learning benchmark to determine the common node set; For each node in the common node set, calculate the difference value of the node's basic attributes as the node's deviation degree, and calculate the difference value of the weight of the edge that the node participates in as the edge deviation penalty value. The node deviation is obtained by weighted summing of the node deviation and the edge deviation penalty value. The consumable categories corresponding to nodes with a comprehensive deviation greater than a preset threshold are the key consumable categories to be optimized. Evaluation results are generated based on the aforementioned key consumable categories.

9. The method according to claim 1, characterized in that, The medical execution unit is a single clinical department, a group of surgeons, or a treatment team.

10. A DRG-based medical consumables usage evaluation system, characterized in that, include: The acquisition module is used to acquire historical medical consumables usage data of patients within the target DRG group; The module is used to construct a consumable usage behavior map for each medical execution unit based on the medical consumable usage data. The scoring module is used to evaluate each medical execution unit based on a multi-dimensional evaluation vector of cost, quality and efficiency dimensions, and select medical execution units whose comprehensive scores meet preset conditions to build a set of benchmark units. The matching module is used to identify medical execution units that are not included in the set of benchmark units as units to be optimized, and to match each unit to be optimized with the benchmark unit with the highest patient profile similarity as a learning benchmark. The analysis module is used to compare and analyze the consumable usage behavior map of the unit to be optimized with the corresponding learning benchmark, identify the key consumable categories to be optimized, and generate evaluation results.

Images