Medical expense insurance simulation decision method and system based on data fusion

Abstract

The application discloses a medical expense insurance simulation decision method and system based on data fusion, and relates to the technical field of medical expense insurance simulation decision. When a target patient has obtained a current stage treatment scheme, but the system cannot determine part of the medication sites to be determined as a unique medication direction according to existing diagnosis information and completed examination results, the candidate medication direction is reserved first, then the supplementary candidate is formed according to the medication category, stage continuation state and subsequent processing path, and the supplementary result of the previous stage corresponding to the same to-be-supplemented site is combined to perform an overlay determination, and only the unoverlaid part generates the supplementary result of this round; when subsequent new data arrives, only the affected unoverlaid part that satisfies the boundary condition and no longer establishes the original support relationship is executed for local reopening, so that the current stage input that can be called by insurance simulation decision is formed.

Description

Technical Field

[0001] This invention relates to the field of medical expense insurance simulation decision-making technology, and more specifically, to a medical expense insurance simulation decision-making method and system based on data fusion. Background Technology

[0002] Medical expense insurance simulation decision-making technology is often used in the long-term treatment process for target patients. Based on the current stage's diagnostic information, completed examination results, and current medication schedule information, it organizes the processing results of subsequent medication items and related medical expenses, forming the current stage input that can be used for subsequent cost calculation and insurance simulation decision-making. In this type of application, existing technologies typically first combine the diagnostic and examination evidence reached at the current stage to screen and converge on various medication decision positions in the current stage's medication schedule; for medication directions that can be directly excluded or directly determined, existing technologies can generally complete the corresponding processing and output the current stage results.

[0003] The existing technology has the following shortcomings: On the one hand, when some medication decision points cannot converge uniquely at the current stage, existing technologies often directly compress them into a single medication direction to quickly form the input for the current stage. However, during long-term treatment, some diagnostic, examination, or treatment pathway evidence may arrive after the current stage's deadline. If the evidence used for the initial compression is insufficient, the previously formed treatment results are prone to becoming invalid after the arrival of new evidence, leading to the restart of the corresponding subsequent treatment chain. As a result, although the current stage input can be formed relatively early, it is prone to repeated rewriting during subsequent updates, making it difficult to maintain consistency in the treatment approach. The difficulty lies in the fact that existing technologies lack an intermediate retention processing layer for situations where evidence is not yet complete, making it difficult to form usable interim results without premature erroneous convergence.

[0004] On the other hand, when existing technologies do not directly compress into a single direction but retain multiple candidate drug directions, each candidate drug direction is usually expanded into independent processing results. While this approach avoids premature compression, it directly increases the number of results in the current round along with the number of candidate directions. Furthermore, during subsequent updates, it is easy to repeatedly expand, compare, and pass the same or equivalent subsequent processing paths, leading to a simultaneous expansion of the result generation and update ranges. As a result, the current input formation process is prone to excessive repetitive generation, redundant processing chains, and difficulty in converging the subsequent update range. The challenge lies in the lack of effective intermediate compression processing between the candidate drug direction layer and the result generation layer, and the failure to continue utilizing the corresponding results formed in the previous stage to shrink the generation range of the current stage.

[0005] To overcome the aforementioned deficiencies of the prior art, embodiments of the present invention provide a data fusion-based medical expense insurance simulation decision-making method to address the problems raised in the background section. Summary of the Invention

[0006] To achieve the above objectives, the present invention provides the following technical solution: A data fusion-based simulation-based decision-making method for medical expense insurance includes: S101: Obtain the diagnostic information, completed examination results, and current medication arrangement information of the target patient in the current stage. Divide the current medication arrangement information into several medication sites to be determined. For each medication site to be determined, filter out the excluded medication directions based on the existing diagnostic information and completed examination results. When a medication site to be determined still retains two or more candidate medication directions and the existing diagnostic information and completed examination results are insufficient to exclude the other candidate medication directions, determine the medication site to be determined as a supplementary site and retain the corresponding candidate medication directions. S102, for each site to be supplemented, the corresponding candidate drug use direction is mapped into a branch difference group according to drug use category, stage continuation status and subsequent treatment path. The candidate drug use directions mapped to the same branch difference group are merged into the same supplementary candidate. Each supplementary candidate is compared with the previous stage supplementary result corresponding to the same site to be supplemented. Only when the previous stage supplementary result is consistent with the current supplementary candidate in drug use category, stage continuation status and subsequent treatment path, and the supporting relationship on which the previous stage supplementary result is based still holds in the current stage, it is determined that the previous stage supplementary result covers the current supplementary candidate, and no new supplementary result is generated for the supplementary candidate in this round. Supplementary results are generated for the supplementary candidates that are not covered, and the supplementary result in this round is determined as the supplementary result in the current stage. S103, detect new evidence arriving at the site to be supplemented. When the new evidence hits the boundary item corresponding to the same site to be supplemented and reopens the mapping table, and causes the supporting relationship on which the supplementation result in the previous stage is no longer valid, only the uncovered supplementation candidates in the corresponding site to be supplemented are partially reopened and the supplementation result is updated. The updated supplementation result is output as the input for insurance simulation decision.

[0007] In a preferred embodiment, the current stage is defined by the target patient's current treatment stage identifier and the stage's cutoff time. Completed examination results are those that have been returned before the stage's cutoff time and have a medication determination conclusion. The medication decision positions in the current stage's medication arrangement information that can independently affect the convergence results of subsequent medication items are divided, with each medication decision position corresponding to a medication site to be determined. Based on the current stage treatment plan description, medication rule relationships, and the correspondence between diagnostic examinations and medication for the target patient corresponding to the medication site to be determined, a set of candidate medication directions corresponding to the medication site to be determined is formed.

[0008] In a preferred embodiment, for any candidate medication direction, the applicable diagnostic conditions, applicable examination conditions, and contraindication exclusion conditions corresponding to the candidate medication direction are read; when the existing diagnostic information and the results of the completed examinations show that the applicable diagnostic conditions are not met, or the applicable examination conditions are not met, or the contraindication exclusion conditions are met, the candidate medication direction is determined to be an excluded medication direction.

[0009] In a preferred embodiment, after the exclusion process is completed, if there are still two or more candidate medication directions for the same medication site to be determined, and at least one diagnostic criterion or examination criterion required to distinguish each candidate medication direction is in a state of not being reached, not being confirmed, or not being sufficient to form a unique convergence conclusion, then the medication site to be determined is identified as a site to be supplemented; a supplementation identifier is generated for the site to be supplemented, and the site identifier associated with the site to be supplemented, the set of retained candidate medication directions, and the criterion items that have not yet met the unique convergence condition are written into the supplementation results of the current stage.

[0010] In a preferred embodiment, within the same supplementary site, the drug category identifier, stage continuation status identifier, and subsequent processing path identifier corresponding to each candidate drug direction are extracted. Candidate drug directions with the same three identifiers are grouped into the same branch difference group. The drug category, stage continuation status, and subsequent processing path corresponding to the branch difference group are retained as a unit. The set of candidate drug directions grouped into the branch difference group is used as the set of directions corresponding to the same supplementary candidate.

[0011] In a preferred embodiment, the supporting relationship upon which the previous stage's supplementary results are based is the correspondence between the diagnostic criteria, examination criteria, and treatment path triggering criteria recorded in the previous stage's supplementary results and the corresponding medication category, stage continuation status, and subsequent treatment path. When the diagnostic information and examination results reached in the current stage do not negate the aforementioned criteria, and there are no new criteria in the current stage that would rewrite the current supplementary candidate as another medication category, another stage continuation status, or another subsequent treatment path, it is determined that the supporting relationship still holds in the current stage. For supplementary candidates that meet the aforementioned conditions, the supplementary candidate identifier and supporting criteria identifier corresponding to the previous stage's supplementary results are retained. For supplementary candidates that are not covered, a supplementary result for this round containing the identifier of the site to be supplemented, the supplementary candidate identifier, the medication category, the stage continuation status, the subsequent treatment path, and the supporting criteria identifier is generated. The retained result and the supplementary result for this round are then merged into the supplementary result for the current stage.

[0012] In a preferred embodiment, for each site to be supplemented, the diagnostic update items, inspection supplement items, and processing path update items that arrive later in the current stage are received, and the items associated with the corresponding site to be supplemented are identified as new additions.

[0013] In a preferred embodiment, the boundary item reopening mapping table records at least the identifier of the site to be filled, the identifier of the boundary item, the type of new basis that can trigger reopening, and the range of affected supplementary candidates; when the basis type, boundary item identifier, and the identifier of the site to be filled corresponding to the new basis are consistent with the records in the boundary item reopening mapping table, it is determined that the new basis hits the boundary item reopening mapping table corresponding to the site to be filled.

[0014] In a preferred embodiment, when the newly added evidence negates at least one of the diagnostic evidence, examination evidence, or treatment path triggering evidence items upon which the previous supplementary result depends, or causes the original evidence combination to no longer support the medication category, stage continuation status, or subsequent treatment path corresponding to the original supplementary candidate, the supporting relationship is determined to be no longer valid. Supplementary results are regenerated only for the affected uncovered supplementary candidates in the sites to be supplemented where the boundary item is reopened and the supporting relationship is no longer valid. For supplementary candidates where the boundary item is not reopened or the supporting relationship is still valid, the original supplementary results are continued to be used. The regenerated supplementary results are merged with the original supplementary results that are continued to be used to obtain the updated supplementary results, and the insurance simulation decision input containing the site to be supplemented identifier, supplementary candidate identifier, current effective medication category, current effective stage continuation status, current effective subsequent treatment path, and current effective supporting evidence identifier is output.

[0015] The data fusion-based medical expense insurance simulation decision-making system includes: a supplementary site identification and candidate retention unit, a supplementary candidate merging and coverage determination unit, a new basis triggering local reopening and input update unit, and a rule relationship and judgment caliber storage area; the supplementary site identification and candidate retention unit, the supplementary candidate merging and coverage determination unit, and the new basis triggering local reopening and input update unit are respectively used to execute S101, S102, and S103; the rule relationship and judgment caliber storage area is used to store and be called by the above units the stage treatment plan description, medication rule relationship, correspondence between diagnosis and examination and medication, supplementary candidate merging caliber, supporting relationship judgment caliber, boundary item reopening registration caliber, new basis hit judgment caliber, and new basis type caliber.

[0016] This invention provides a data fusion-based medical expense insurance simulation decision-making method and system, and its effects and advantages are as follows: This invention provides a medical expense insurance simulation decision-making method and system based on data fusion. For medication sites that cannot converge uniquely, it first retains candidate medication directions, then forms supplementary candidates according to the dimensions that truly affect the differences in subsequent processing chains. It then combines the previous stage's supplementary results corresponding to the same medication site to perform coverage determination. This transforms the current stage input formation process from a single compression or separate expansion in each direction when evidence is insufficient to a consistent processing approach of first retaining, then compressing, and finally generating, thereby reducing the re-expansion of subsequent processing chains caused by premature erroneous convergence and reducing repeated expansion, comparison, and transmission of similar medication branches. Furthermore, when new evidence arrives, this invention only performs partial reopening on affected, uncovered supplementary candidates that meet boundary conditions and whose original supporting relationships no longer hold. This transforms the update process from overall recalculation to targeted updates within the affected range, thereby improving the consistency and stability of the current stage input formation and update. It also provides a more stable basis for subsequent insurance simulation decisions without changing the existing diagnostic information, completed examination results, and current stage medication arrangement information input foundation. Attached Figure Description

[0017] Figure 1 This is a schematic diagram of the method flow of the present invention; Figure 2 This is a schematic diagram of the system structure of the present invention; Figure 3 This is a schematic diagram illustrating the candidate drug direction merging relationship of the present invention; Figure 4 A schematic diagram illustrating the expansion of supplementary results under traditional processing; Figure 5 This is a schematic diagram of the overlay reuse and partial reopening processing of the present invention. Detailed Implementation

[0018] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0019] This invention provides a medical expense insurance simulation decision-making method and system based on data fusion. In a long-term treatment expense insurance simulation scenario targeting a patient, the system typically needs to form the current stage input, which can be used for subsequent insurance simulation decisions, based on the current stage treatment plan and combined with the current stage diagnostic information and completed examination results.

[0020] In this scenario, existing approaches typically begin by filtering out excluded medication directions from each potential medication site in the current treatment plan. When a potential medication site cannot be directly identified as the only possible medication direction, it is usually either compressed into a single medication direction or multiple candidate medication directions are retained and supplemented separately. The default premise of this approach is that the existing evidence at the current stage is sufficient to support the stable determination of the subsequent processing chain, or that expanding each direction will not significantly amplify the cost of supplementary generation and subsequent updates.

[0021] In actual operation, some sites awaiting medication determination may still retain two or more candidate medication directions based on existing diagnostic information and completed examination results. Furthermore, at least one diagnostic or examination criterion required to distinguish these candidate medication directions may still be unreached, unconfirmed, or insufficient to form a unique convergent conclusion. In this case, the aforementioned default premise no longer holds. If the results are still directly compressed into a single medication direction, the arrival of subsequent new criterion may invalidate the original compressed results, causing the corresponding subsequent processing chain to be re-expanded. If each candidate medication direction is still expanded into independent supplementary results, the number of supplementary results in this round will directly expand with the number of candidate medication directions, causing similar medication branches to be repeatedly expanded, compared, and transmitted.

[0022] Therefore, the technical problem this invention aims to solve is, under the constraint that the evidence at the current stage is insufficient to uniquely converge the drug site to be determined, how to first retain candidate drug directions, then form supplementary candidates according to the dimensions that truly affect the differences in subsequent processing chains, and perform coverage determination in conjunction with the previous stage supplementary results corresponding to the same drug site to be determined. Only supplementary results for this round are generated for supplementary candidates that are not covered. Simultaneously, when new evidence arrives later, only affected and uncovered supplementary candidates are partially reopened, thereby generating the current stage input that can be used for insurance simulation decision-making. To this end, this invention organizes the current stage input formation process around drug sites that cannot uniquely converge, generating insurance simulation decision input through candidate drug direction retention, supplementary candidate merging, previous stage result coverage determination, and partial reopening triggered by boundary conditions.

[0023] Based on the above design, this invention constructs a complete process for a data fusion-based medical expense insurance simulation decision-making method, consisting of steps S101 to S103 sequentially. (Refer to...) Figure 1 , Figure 1 This is a schematic diagram of the method flow of the present invention, which includes: Step S101, "Identification and Candidate Retention of Supplementary Sites," is used when the current treatment plan for the target patient has been established, but existing evidence is insufficient to uniquely converge the treatment site to be determined. It identifies the treatment site to be determined and retains candidate medication directions, preparing the necessary entry points for subsequent supplementary candidate merging and coverage determination. This step reads the current stage diagnostic information X101, completed examination results X102, and current stage medication arrangement information X103. It performs site splitting, candidate medication direction formation, exclusion of excluded medication directions, and determination of supplementary sites in the current stage medication arrangement information. The supplementary site identifier, the set of retained candidate medication directions, the basis items that have not yet met the unique convergence condition, and the current stage identifier are written into the current stage supplementary result R101, for downstream step S102 to read and execute supplementary candidate merging and coverage determination. X103 can be directly derived from the current stage treatment plan determined by the doctor, or it can be derived from the current stage treatment plan generated by the system based on the target patient's disease information and by calling the clinical knowledge database. The current stage supplementary result R101 generated in this step provides the supplementary site entry and candidate retention basis for the subsequent step S102 to generate supplementary candidates and perform coverage determination, and provides a unified reading standard for the generation of the current stage supplementary result.

[0024] Step S102, supplementary candidate merging and coverage determination, is used to form supplementary candidates based on the drug category, stage continuation status, and subsequent processing path, after the candidate retention of the site to be supplemented has been completed. It also combines the supplementary results of the previous stage to perform coverage determination and boundary item registration, forming the current stage supplementary results for subsequent partial reopening and input update. This step reads the current stage supplementary result R101 and the previous stage supplementary result X104, performs branch difference group formation, supplementary candidate merging, coverage determination, and boundary item reopening mapping table registration for the candidate drug directions retained within the same site to be supplemented. The current stage supplementary results, whether used or newly generated, are written into R102. The boundary item identifier, the new basis type that can trigger reopening, the affected supplementary candidate range, and the corresponding original support relationship reference identifier are written into R103, for downstream step S103 to read and perform partial reopening and input update. The supplementary result R102 generated in this step provides subsequent step S103 with supplementary candidate results that are either used in the current stage or newly generated, along with their supporting basis. The boundary item reopening mapping table R103 generated in this step provides subsequent step S103 with the boundary triggering range and local reopening basis after the new basis is hit. After the coverage determination is completed in this stage, the supplementary result R102 not only serves as the basis for the results of local reopening and input update in step S103, but also serves as the source basis for the supplementary results of the previous stage in subsequent stages after the stage progresses.

[0025] Step S103 adds new evidence to trigger partial reopening and input update. This step determines whether new evidence hits the boundary item reopening condition and causes the original supporting relationship to fail after subsequent new evidence arrives. For affected and uncovered supplementary candidates, partial reopening is performed, forming the current stage's valid input for subsequent treatment cost calculation and insurance simulation decision-making. This step reads the current stage supplementary result R102, the boundary item reopening mapping table R103, and subsequent new evidence X105. For new evidence associated with the site to be supplemented, it performs hit determination, supporting relationship failure determination, partial reopening range limitation, and supplementary result update. The updated current valid medication category, current valid stage continuation status, current valid subsequent processing path, current valid supporting evidence identifier, and update source identifier are written into the insurance simulation decision input R104. The insurance simulation decision input R104 formed in this step provides the current valid supplementary result and evidence source after partial reopening for subsequent treatment cost calculation and insurance simulation decision-making. It retains the results of unaffected supplementary candidates and updates supplementary candidates that hit the boundary condition and whose supporting relationship fails to the current valid result after reopening.

[0026] For ease of understanding, the following embodiments are described under a unified system architecture, which can be modified equivalently according to actual needs. The system consists of a unit for identifying and retaining candidate sites, a unit for merging supplementary candidates and determining coverage, a unit for triggering local reopening and input updating based on new additions, and a storage area for rule relationships and determination criteria. (Refer to...) Figure 2 , Figure 2 This is a schematic diagram of the system structure of the present invention.

[0027] Specifically, the rule relationship and judgment caliber storage area is used to store the treatment plan description, medication rule relationship, correspondence between diagnosis and examination and medication, supplementary candidate merging caliber, supporting relationship judgment caliber, boundary item reopening registration caliber, newly added basis hit judgment caliber, and newly added basis type caliber, and is available for reading by the aforementioned three units respectively.

[0028] The supplementary site identification and candidate retention unit is used to execute method step S101. It receives the current stage diagnosis information X101, the completed examination results X102 and the current stage medication arrangement information X103, reads the stage treatment plan description, medication rule relationship and the correspondence between diagnosis and examination and medication, generates the current stage supplementary result R101 and outputs it.

[0029] The supplementary candidate merging and coverage determination unit is used to execute method step S102. It receives the current stage supplementary result R101 and the previous stage supplementary result X104, reads the supplementary candidate merging caliber, the support relationship determination caliber, and the boundary item reopening registration caliber, generates the current stage supplementary result R102 and the boundary item reopening mapping table R103, and outputs them.

[0030] The newly added basis triggers partial reopening and input update unit is used to execute method step S103. It receives the current stage supplementary result R102, the boundary item reopening mapping table R103 and the subsequent newly added basis X105, reads the support relationship judgment caliber, the newly added basis hit judgment caliber and the newly added basis type caliber, generates insurance simulation decision input R104 and outputs it.

[0031] Therefore, the system provides each unit with the preset relationships and judgment criteria required for the corresponding steps through the rule relationship and judgment caliber storage area, and each unit executes the method steps S101 to S103 in sequence to complete the identification of the site to be supplemented, the supplementary candidate merging and coverage judgment, and the local reopening and input update triggered by the new basis.

[0032] In an optional embodiment, the implementation process of step S101 includes information acquisition and site segmentation, candidate drug direction formation and screening, determination of sites to be supplemented and result writing back.

[0033] In the information acquisition and site segmentation process, the current stage diagnostic information X101, completed examination results X102, and current stage medication arrangement information X103 are read first. The current stage is defined by the target patient's current treatment stage identifier and the corresponding stage cutoff time. The completed examination results X102 are examination results that have been returned before the stage cutoff time and have a medication determination conclusion.

[0034] It should be noted that when X103 is directly derived from the doctor's established current stage treatment plan, the system extracts the medication arrangement items related to the current stage from the target patient's corresponding stage medical order record, treatment plan record, or prescription record, and converts the medication category, stage continuation status, and subsequent processing path into a unified field format. When the current stage treatment plan has not yet been fully provided by the doctor, the system receives the target patient's disease information, current treatment stage identifier, and previous examination results, and uses this information to access the clinical knowledge database to filter candidate stage treatment plans that meet the current constraints. The knowledge sources of the clinical knowledge database include at least clinical practice guidelines, clinical pathways, technical operation specifications, drug instructions, expert consensus, and medical literature, and based on the aforementioned knowledge sources, it extracts rule relationships between disease information, examination conditions, indications, contraindications, and stage treatment plans. The system extracts the medication arrangement items corresponding to the current stage from the selected candidate stage treatment plans and converts them into current stage medication arrangement information X103.

[0035] Subsequently, the medication decision positions in the current medication arrangement information X103 that can independently affect the convergence results of subsequent medication items are split, so that each medication decision position corresponds to a medication site to be determined, which serves as the basic processing object for the formation and exclusion of subsequent candidate medication directions.

[0036] In the formation and screening of candidate medication directions, for each site to be determined for medication, the stage treatment plan description, medication rule relationship, and correspondence between diagnostic examinations and medication associated with that site are read, and an initial set of candidate medication directions corresponding to that site is enumerated accordingly. Specifically, the stage treatment plan description records the treatment arrangement items related to the site to be determined in the current stage; the medication rule relationship records the rule correspondence between applicable diagnostic conditions, applicable examination conditions, contraindication exclusion conditions, and candidate medication directions; and the correspondence between diagnostic examinations and medication records the mapping relationship between diagnostic items, examination items, and candidate medication directions. For any candidate medication direction, its corresponding applicable diagnostic conditions, applicable examination conditions, and contraindication exclusion conditions are extracted and evaluated item by item in conjunction with the current stage diagnostic information X101 and the completed examination results X102. If the current stage diagnostic information X101 indicates that the applicable diagnostic conditions corresponding to the candidate medication direction are not met, then the candidate medication direction is determined to be excluded. If the completed examination result X102 indicates that the applicable examination conditions for the candidate medication direction are not met, then the candidate medication direction is determined to be excluded. If the existing diagnostic information X101 and the completed examination result X102 indicate that the candidate medication direction meets the corresponding contraindication exclusion conditions, then the candidate medication direction is also determined to be excluded. After the above screening, the candidate medication directions that have not been excluded are retained, resulting in the set of remaining candidate medication directions corresponding to each medication site to be determined.

[0037] In the process of determining and writing back results for sites to be supplemented, a unique convergence determination is performed on each site to be supplemented after screening. If a site to be supplemented has only one candidate medication direction after screening, it is determined that the site has achieved unique convergence in the current stage and will not be written into the current stage's supplementary result R101. If a site to be supplemented still retains two or more candidate medication directions after screening, and at least one diagnostic or examination criterion required to distinguish the candidate medication directions is in a state of not being reached, not being confirmed, or insufficient to form a unique convergence conclusion, it is determined that the site to be supplemented has not achieved unique convergence in the current stage and is identified as a site to be supplemented. For objects identified as sites to be supplemented, a supplementation identifier is generated for them, and the site identifier associated with the site to be supplemented, the supplementation identifier, the set of retained candidate medication directions, the criterion items that have not yet met the unique convergence condition, and the current stage identifier are written into the current stage's supplementary result R101. The current stage result R101 carries the field of the site to be supplemented, the field of the supplemented identifier, the field of the candidate drug direction set, the field of the basis for not meeting the unique convergence condition, and the current stage identifier. The output of R101 in this step allows the subsequent step S102 to directly perform supplementary candidate merging and coverage determination based on this structure without having to re-execute site splitting and direction screening.

[0038] In an optional embodiment, after reading the current stage supplementary result R101 and the previous stage supplementary result X104, step S102 forms supplementary candidates based on the drug category, stage continuation status, and subsequent processing path for the candidate drug directions retained within the same supplementary site range. It then performs a coverage determination based on the previous stage supplementary result to generate the current stage supplementary result R102. Simultaneously, it registers the boundary item reopening mapping table R103, thus providing a result basis and boundary reference for the subsequent local reopening and input update in step S103. The implementation process of step S102 includes four parts: branch difference group formation, supplementary candidate formation, coverage determination and writing back the current stage supplementary result, and boundary item reopening mapping table registration.

[0039] In the formation of branch difference groups, the set of retained candidate drug directions corresponding to each site to be supplemented in R101 is read. Within the same site to be supplemented, the drug category identifier, stage continuation status identifier, and subsequent processing path identifier corresponding to each candidate drug direction are extracted. Candidate drug directions with the same three identifiers are grouped into the same branch difference group. The branch difference group is used to carry the group-level processing scope when merging candidate drug directions for supplementary candidates under the same site to be supplemented. Its processing scope is limited to within the same site to be supplemented and does not merge across sites to be supplemented. The supplementary candidate merging scope includes at least three merging dimensions: drug category, stage continuation status, and subsequent processing path. Within the same site to be supplemented, only when the three merging dimensions are consistent can the corresponding candidate drug directions be grouped into the same branch difference group and continue to form the same supplementary candidate. Thus, objects that correspond to different candidate drug directions under the same site to be supplemented but do not produce differences in subsequent processing chains in terms of drug category, stage continuation status, and subsequent processing path can be compressed into the same branch difference group, providing a group-level merging basis for the formation of subsequent supplementary candidates.

[0040] In the formation of supplementary candidates, the corresponding medication category, stage continuation status, and subsequent processing path are retained for each branch difference group. The set of candidate medication directions belonging to that branch difference group is taken as the direction set corresponding to the same supplementary candidate, thus forming supplementary candidates under the current stage's site to be supplemented. For multiple candidate medication directions within the same site to be supplemented, if they are identical in the three identifiers of medication category, stage continuation status, and subsequent processing path, it indicates that although these candidate medication directions correspond to different upstream support sources, they are consistent in the dimension that truly determines the difference in the subsequent processing chain at the current stage, and therefore can be processed as the same supplementary candidate. Thus, each supplementary candidate is associated with at least the site to be supplemented identifier, supplementary candidate identifier, the direction set corresponding to the supplementary candidate, medication category, stage continuation status, and subsequent processing path. For each supplementary candidate formed for the same site to be supplemented, subsequent processing is no longer carried out separately for each candidate medication direction, but rather coverage determination and result generation are performed on a supplementary candidate basis.

[0041] In the coverage determination and current stage supplementary result write-back process, each supplementary candidate is compared with the previous stage supplementary result corresponding to the same supplementary site. The previous stage supplementary result is deemed to cover the current supplementary candidate only if it matches the current supplementary candidate in medication category, stage continuation status, and subsequent treatment path, and the supporting relationship upon which the previous stage supplementary result is based still holds in the current stage. The criteria for determining the supporting relationship include the composition criteria and the criteria for whether the supporting relationship still holds. The supporting relationship of the previous stage supplementary result refers to the correspondence between the diagnostic criteria, examination criteria, and treatment path triggering criteria recorded in the previous stage supplementary result, and the corresponding medication category, stage continuation status, and subsequent treatment path. If the diagnostic information and examination results reached in the current stage do not negate the aforementioned criteria, and there are no new criteria in the current stage that would rewrite the current supplementary candidate to another medication category, another stage continuation status, or another subsequent treatment path, the supporting relationship is deemed to still hold in the current stage.

[0042] When the coverage determination is valid, the previous stage's supplementary results are retained, and no new supplementary results are generated for this round for the supplementary candidate. The coverage status of the supplementary candidate is recorded as covered. When the coverage determination is invalid, a new supplementary result is generated for the supplementary candidate, and the coverage status of the supplementary candidate is recorded as uncovered. An invalid coverage determination includes two scenarios: first, the previous stage's supplementary results and the current supplementary candidate are inconsistent in terms of medication category, stage continuation status, or subsequent processing path; second, the supporting relationship no longer holds in the current stage. For uncovered supplementary candidates, a supplementary result record applicable to the current stage can be generated based on their corresponding medication category, stage continuation status, subsequent processing path, and the supporting evidence already reached in the current stage. The newly generated supplementary result for this round must at least include the supplementary candidate identifier, medication category, stage continuation status, subsequent processing path, and current stage supporting evidence identifier. After the above processing, the retained results corresponding to the covered supplementary candidate are merged with the newly generated results corresponding to the uncovered supplementary candidate to obtain the current stage supplementary result R102.

[0043] The current supplementary result R102 at least carries the identifier of the site to be supplemented, the supplementary candidate identifier, the direction set corresponding to the supplementary candidate, the medication category, the stage continuation status, the subsequent processing path, the coverage status, the supporting evidence identifier, and the identifier of the continued result or the newly generated result. Among them, the coverage status is used to distinguish whether the supplementary candidate is using the result of the previous stage or generating a new result in this round in the current stage; the supporting evidence identifier is used to record the source of the evidence supporting the validity of the current supplementary result; and the continued result identifier or the newly generated result identifier is used to identify the source of the formation of the supplementary result, so that the subsequent step S103 can directly distinguish the current valid status and result source of each supplementary candidate when reading R102.

[0044] Therefore, the supplementary result R102 in the current stage carries the continued results and newly generated results of the supplementary candidates in this stage, and serves as the source of inheritance for the supplementary results of the previous stage in subsequent stages after the stage progresses. This allows the same site to be supplemented to be recursively pushed along the result chain between adjacent stages, rather than re-expanding all supplementary generation in each stage.

[0045] In the boundary item reopening mapping table registration, for supplementary candidates that are not covered under each supplementary site and whose reopening boundary constraints need to be retained, boundary items that can cause the original support relationship to fail and trigger subsequent local reopening are extracted and registered to form the boundary item reopening mapping table R103. The boundary item reopening registration criteria include at least the boundary item identifier, the type of new basis that can trigger reopening, the range of affected supplementary candidates, and the registration method of the corresponding original support relationship reference identifier. The boundary item is used to characterize the key basis position that is directly related to the establishment boundary of the supplementary candidate and may change the current establishment state of the support relationship after the arrival of subsequent new basis. The boundary item identifier may correspond to at least one basis position identifier that plays a decisive role in the establishment boundary of the current supplementary candidate among diagnostic basis items, inspection basis items, or processing path triggering basis items. For each boundary item registration record, at least the supplementary site identifier, boundary item identifier, type of new basis that can trigger reopening, range of affected supplementary candidates, and corresponding original support relationship reference identifier are written. The newly added basis type that can trigger reopening is used to limit which subsequent diagnostic update items, check supplementary items, or processing path update items can trigger the local reopening corresponding to the boundary item. The affected supplementary candidate range is used to limit the supplementary candidate objects that need to be re-checked after hitting the boundary item. The affected supplementary candidate range is at least used to record the set of supplementary candidate identifiers that need to be re-checked after hitting the boundary item, or a range identifier that can uniquely locate the supplementary candidate set. The corresponding original support relationship reference identifier is used to associate the location of the support basis record or support relationship record corresponding to the supplementary candidate in R102.

[0046] Therefore, while generating the supplementary results for the current stage, step S102 pre-establishes the boundary trigger range and the location relationship of the affected objects after the arrival of the subsequent new basis, so that step S103 can determine whether the new basis hits the reopening condition according to the identifier of the site to be supplemented, the basis type and the boundary item identifier, and only perform local reopening on the supplementary candidates that hit the boundary condition and whose support relationship is invalid, without having to recalculate all the supplementary results as a whole.

[0047] In an optional embodiment, after reading the current stage supplementary result R102, the boundary item reopening mapping table R103, and the subsequent newly added evidence X105, step S103 sequentially performs boundary item hit determination, original support relationship failure determination, local reopening range limitation, and supplementary result update for the newly added evidence associated with the site to be supplemented, generating insurance simulation decision input R104. This ensures that when subsequent newly added evidence arrives, only affected and uncovered supplementary candidates are partially reopened, while the results of unaffected supplementary candidates are retained. The implementation process of step S103 includes three parts: new evidence association and hit determination, support relationship failure determination and local reopening range determination, and supplementary result update and insurance simulation decision input write-back.

[0048] In the addition of evidence association and hit determination, subsequent additions X105 are read, and diagnostic update items, examination addition items, and treatment path update items associated with each site to be supplemented are identified as new evidence to be determined. The criteria for the type of new evidence include at least the classification methods for diagnostic update items, examination addition items, and treatment path update items. The association can be established based on the site to be supplemented identifier, the current stage identifier of the target patient, and the association relationship between the new evidence and the original evidence items, so that the new evidence is limited to the corresponding site to be supplemented before entering the hit determination. The association relationship between the new evidence and the original evidence items can be determined by the correspondence relationship of evidence item identifiers, site reference relationships, or supporting evidence identifier reference relationships. The criteria for new evidence hit determination include at least the consistency comparison method of the site to be supplemented identifier, evidence type, and boundary item identifier. Subsequently, in combination with the site to be supplemented identifier, boundary item identifier, and new evidence type that can trigger reopening registered in R103, a hit determination is performed on each new evidence to be determined. When the basis type, boundary item identifier, and the identifier of the site to be supplemented corresponding to the newly added basis match the registration record in R103, it is determined that the newly added basis hits the boundary item reopening mapping table corresponding to the site to be supplemented, and it is marked as a newly added basis that hits the boundary item, proceeding to the subsequent support relationship failure determination. For newly added basis that does not hit the boundary item reopening mapping table, the corresponding supplementary candidate's local reopening process is not triggered. Thus, the newly added basis that arrives later can be limited to the hit objects that need to be further checked and the non-hit objects that do not need to trigger local reopening, providing a preliminary screening basis for the subsequent convergence of the reopening range.

[0049] In determining the failure of supporting relationships and the scope of partial reopening, for newly added evidence that hits the reopening condition of the boundary item, the supporting evidence identifier, coverage status, and retained result identifier or newly generated result identifier associated with the corresponding supplementary candidate in R102 are read to determine whether the original supporting relationship is no longer valid due to the newly added evidence. The original supporting relationship no longer being valid means that the newly added evidence negates at least one of the original diagnostic evidence, original examination evidence, or original treatment path triggering evidence, or makes the original evidence combination no longer support the medication category, stage continuation status, or subsequent treatment path corresponding to the current supplementary candidate. For supplementary candidates that hit the boundary item and whose original supporting relationship is no longer valid, the scope of affected supplementary candidates registered in R103 is further read to determine the objects of partial reopening. The scope of partial reopening is limited to supplementary candidates that simultaneously meet the following conditions: hitting the boundary item reopening condition, the original supporting relationship no longer being valid, and the coverage status in R102 is uncovered. For supplementary candidates that do not hit the boundary item, or supplementary candidates that hit the boundary item but whose original supporting relationship is still valid, the current valid results in R102 are continued, and they are not entered into the partial reopening process. For supplementary candidates whose coverage status is already covered, even if new criteria hit the boundary item, they will remain in the current processing state and will not be included in the local reopening object set. Therefore, the processing scope after subsequent new criteria arrive can be narrowed down to uncovered supplementary candidate objects that hit the boundary condition and whose supporting relationship has failed, instead of expanding to all supplementary results.

[0050] In the supplementary result update and insurance simulation decision input write-back process, for supplementary candidates entering the partial reopening scope, based on the updated diagnostic criteria, examination criteria, or processing path trigger criteria, the current effective medication category, current effective stage continuation status, and current effective subsequent processing path corresponding to the supplementary candidate are redefined, and its current effective supporting criteria identifier is updated. The current effective result indicates the status of the supplementary result that can be directly used for insurance simulation decision invocation after the arrival of the newly added criteria and the completion of the partial reopening process. For supplementary candidates that have not entered the partial reopening scope, the current effective result in R102 is continued to be used and not regenerated. After the above processing, the result regenerated after partial reopening is merged with the original result that is continued to be used to form the current stage effective input R104 for insurance simulation decision invocation.

[0051] The insurance simulation decision input R104 must include at least the identifier of the site to be supplemented, the currently effective drug category, the current effective stage continuation status, the current effective subsequent processing path, the currently effective supporting evidence identifier, and the update source identifier. The current effective supporting evidence identifier records the source of evidence corresponding to the current effective result in this round of output, while the update source identifier distinguishes whether the current effective result uses the original supplementary result from R102 or is regenerated from a partial reopening in this round. Therefore, when downstream insurance simulation decisions read R104, they do not need to backtrack to determine whether each supplementary candidate has undergone partial reopening; they can directly complete the simulation decision invocation for the current stage based on the current effective result and the update source. The current effective result after partial reopening processing constitutes both the expression of the effective stage result after the current stage is updated and provides the result basis for the continued execution of inheritance judgments in subsequent stages, enabling the same site to form a closed-loop chain during stage advancement: inheritance of results from the previous stage, partial correction in the current stage, and continued inheritance in subsequent stages.

[0052] In an optional application embodiment, the insurance simulation decision input R104 can also serve as the entry point for the treatment cost insurance simulation processing for the target patient. Instead of re-executing the full cost expansion for the current stage of the treatment plan, the system first reads the currently effective medication category, the current effective stage continuation status, the currently effective subsequent processing path, the currently effective supporting evidence identifier, and the update source identifier from R104, and determines the currently effective treatment branch that needs to enter the subsequent cost simulation. Thus, treatment branches that have already been used through coverage determination in previous steps continue to be used in downstream cost processing; treatment branches that were partially reopened in previous steps due to newly added evidence hitting boundary conditions enter the subsequent cost simulation with the updated currently effective results, thereby ensuring that the scope of downstream cost processing remains consistent with the effective treatment branches that have converged in the midstream stage of this invention.

[0053] Based on the aforementioned currently effective treatment branches, the system further generates treatment cost calculation entries. These treatment cost calculation entries are structured cost result units formed for the subsequent treatment content actually corresponding to the currently effective treatment branch in R104. They are not a mechanical expansion of all possible treatment items, but only form calculation results for the cost objects corresponding to the currently effective medication category, the current effective stage's continuation status, and the current effective subsequent processing path. Each treatment cost calculation entry corresponds to at least one cost object under the currently effective treatment branch, and records whether the cost object is continued in the current stage, whether it has entered the subsequent processing chain, and the corresponding payment adaptation relationship. For treatment branches whose update source identifier indicates a need for partial reopening, the system regenerates their corresponding treatment cost calculation entries; for treatment branches whose update source identifier indicates continued use, the system continues to use their corresponding treatment cost calculation entries. Therefore, subsequent cost simulations do not require recalculating all cost objects, but only update the local cost range affected by the newly added basis.

[0054] After generating treatment cost calculation entries, the system constructs different insurance scenarios around the same current stage treatment plan and performs cost settlement simulations for each, generating insurance scenario results. These insurance scenario results refer to the set of cost results obtained for the same current stage treatment plan under different basic medical insurance and supplementary insurance conditions. They reflect, at least under basic medical insurance conditions, employee medical insurance conditions, supplementary insurance conditions, or user-input reimbursement conditions, the pre-reimbursement cost, post-reimbursement cost, and remaining out-of-pocket cost for the corresponding currently effective treatment branch. For treatment branches that remain unchanged due to reuse from previous steps, the corresponding cost results under each insurance scenario continue to be used. For treatment branches that partially reopen due to newly added criteria hitting boundary conditions, the system only recalculates the affected cost results under their corresponding insurance scenarios. Therefore, the cost simulation results under different insurance scenarios can be partially corrected as new criteria arrive, without needing to re-execute the entire cost simulation for all current stage treatment plans.

[0055] After completing calculations under different insurance scenarios, the system generates a visualized decision report based on the results. This visualized decision report outputs the cost outcomes under different insurance scenarios in the form of charts, lists, or comparative summaries. It includes at least a summary of the current effective treatment branches, a comparison of pre-reimbursement costs under different insurance scenarios, a comparison of post-reimbursement costs, a comparison of remaining out-of-pocket expenses, and an indicator of changes in results caused by partial reopening triggered by new evidence. Thus, target patients and their families can not only see the cost differences under different insurance scenarios but also identify whether the cost changes in this round stem from the continuation of the original results or from partial updates triggered by new evidence. This allows for comparable, traceable, and continuously updated treatment cost insurance decision-making results, even when subsequent evidence for the treatment plan is still dynamically arriving.

[0056] To illustrate the processing logic of this invention in terms of controlling the range of the supplementary site generation results and subsequent local reopening control, Figures 3 to 5 This simulation sample is constructed based on a method consistent with that of the present invention. This simulation sample is not arbitrarily assembled, but generated according to the object relationships involved in the present invention, and includes at least objects such as the site to be supplemented, candidate drug directions, branch difference groups, previous stage supplementation results, current stage basis status, newly added basis, and boundary reopening mapping relationships. For ease of demonstration, the same site to be supplemented, W001, is extracted from this simulation sample as an example object.

[0057] During the sample construction process, multiple candidate drug application directions are first generated for the site to be supplemented, W001, and each candidate drug application direction is assigned three difference attributes: drug category, stage continuation status, and subsequent processing path. Then, the candidate drug application directions are merged according to these three difference attributes to form branch difference groups. Subsequently, combined with the supplementary results of the previous stage and their supporting relationships, the evidence that has been reached, evidence that has not been reached, and evidence that has not been confirmed in the current stage are generated, and a reopening mapping relationship between the newly added evidence and the boundary items is established to describe the situations where there are mergeable situations, situations where the results of the previous stage can be covered, situations where the results of the previous stage cannot be covered, and situations where only the affected area is triggered by local reopening.

[0058] Figures 3 to 5 In the diagram, D001 to D006 represent candidate drug directions corresponding to the site W001 to be supplemented; G001 to G003 represent branch difference groups formed by merging based on difference attributes; C001 to C003 represent supplementary candidates generated from each branch difference group; PR001 and PR002 represent supplementary results from the previous stage; R001 and R002 represent currently valid results after coverage reuse; CR003 represents newly generated and uncovered supplementary results from the current stage; NE001 represents newly added criteria; B003 represents boundary items; and U003 represents results updated after partial reopening. The above numbering is for illustrative purposes only and does not constitute a limitation on the number of objects, their arrangement, or the processing order.

[0059] Figure 3 This is used to illustrate the merging relationship of candidate drug use directions. For multiple candidate drug use directions under the same site to be supplemented, instead of directly generating independent supplementation results one by one, they are first merged according to drug category, stage continuation status and subsequent processing path to form branch difference groups. Then, each branch difference group generates corresponding supplementation candidates, providing a unified processing unit for subsequent coverage reuse and partial reopening.

[0060] Figure 4This is used to illustrate the traditional processing path. In the traditional processing method, each candidate drug direction retained for the same site to be supplemented generates independent supplementary results, and the supplementary results of the previous stage are not used to compress the range generated in the current stage. Therefore, as the number of candidate drug directions increases, the range of newly generated results in the current stage expands synchronously, resulting in an increase in the number of results and an expansion of the subsequent maintenance scope.

[0061] Figure 5 This describes the processing path of the present invention. In this invention, candidate drug directions are first merged, and then the current stage's supplementary candidates are compared with the previous stage's supplementary results for coverage determination. For covered supplementary candidates, the previous stage's supplementary results are directly reused to form the current valid results. For uncovered supplementary candidates, corresponding supplementary results are generated in the current stage. When a new mapping relationship is reopened based on the hit boundary, causing existing support relationships to fail, only the affected area is partially reopened and the results are updated. Therefore, the present invention can compress the generation range of the current stage's supplementary results and limit subsequent reopenings to necessary local areas.

[0062] The above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented, in whole or in part, in the form of a computer program product.

[0063] Those skilled in the art will recognize that the modules and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and inventive constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0064] In addition, the functional modules in the various embodiments of this application can be integrated into one processing module, or each module can exist physically separately, or two or more modules can be integrated into one module.

[0065] The above description is merely a specific 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 medical expense insurance simulation decision-making method based on data fusion, characterized in that, include: S101, obtain the current stage diagnostic information, completed examination results and current stage medication arrangement information of the target patient, break down the current stage medication arrangement information into several medication sites to be determined, screen out the excluded medication directions for each medication site in combination with the existing diagnostic information and completed examination results, when a certain medication site still retains two or more candidate medication directions and the existing diagnostic information and completed examination results are insufficient to exclude the other candidate medication directions, determine the medication site to be determined as a supplementary site and retain the corresponding candidate medication directions; S102, for each site to be supplemented, the corresponding candidate drug use direction is mapped into a branch difference group according to drug use category, stage continuation status and subsequent treatment path. The candidate drug use directions mapped to the same branch difference group are merged into the same supplementary candidate. Each supplementary candidate is compared with the previous stage supplementary result corresponding to the same site to be supplemented. Only when the previous stage supplementary result is consistent with the current supplementary candidate in drug use category, stage continuation status and subsequent treatment path, and the supporting relationship on which the previous stage supplementary result is based still holds in the current stage, it is determined that the previous stage supplementary result covers the current supplementary candidate, and no new supplementary result is generated for the supplementary candidate in this round. Supplementary results are generated for the supplementary candidates that are not covered, and the supplementary result in this round is determined as the supplementary result in the current stage. S103, detect new evidence arriving at the site to be supplemented. When the new evidence hits the boundary item corresponding to the same site to be supplemented and reopens the mapping table, and causes the supporting relationship on which the supplementation result in the previous stage is no longer valid, only the uncovered supplementation candidates in the corresponding site to be supplemented are partially reopened and the supplementation result is updated. The updated supplementation result is output as the input for insurance simulation decision.

2. The medical expense insurance simulation decision-making method based on data fusion according to claim 1, characterized in that, The current stage is defined by the target patient's current treatment stage identifier and the end time of this stage. Completed examination results are those that have been returned before the end time of this stage and have a medication determination conclusion. The medication decision position in the current stage medication arrangement information that can independently affect the convergence result of subsequent medication items is divided, and each medication decision position corresponds to a medication site to be determined. Based on the description of the current stage treatment plan, medication rules, and the correspondence between diagnostic examinations and medication for the target patient corresponding to the medication site to be determined, a set of candidate medication directions corresponding to the medication site to be determined is formed.

3. The medical expense insurance simulation decision-making method based on data fusion according to claim 2, characterized in that, For any candidate drug direction, read the applicable diagnostic conditions, applicable examination conditions, and contraindication exclusion conditions corresponding to that candidate drug direction; When existing diagnostic information and completed examination results indicate that the applicable diagnostic conditions are not met, or the applicable examination conditions are not met, or the contraindication exclusion conditions are met, the candidate medication direction is determined to be excluded.

4. The medical expense insurance simulation decision-making method based on data fusion according to claim 3, characterized in that, After the exclusion process is completed, if there are still two or more candidate drug directions for the same drug site to be determined, and at least one diagnostic or examination basis required to distinguish each candidate drug direction is in a state of not being reached, not being confirmed, or not being sufficient to form a unique convergence conclusion, then the drug site to be determined is identified as a site to be supplemented. Generate a supplementary identifier for the site to be supplemented, and write the site identifier associated with the site to be supplemented, the set of candidate drug directions to be retained, and the basis items that have not yet met the unique convergence condition into the supplementary results of the current stage.

5. The medical expense insurance simulation decision-making method based on data fusion according to claim 1, characterized in that, Within the same supplementary site, extract the drug category identifier, stage continuation status identifier, and subsequent processing path identifier corresponding to each candidate drug direction. Candidate drug directions with the same three identifiers are grouped into the same branch difference group. The drug category, stage continuation status, and subsequent processing path corresponding to this branch difference group are retained as a unit. The set of candidate drug directions grouped into this branch difference group is used as the set of directions corresponding to the same supplementary candidate.

6. The medical expense insurance simulation decision-making method based on data fusion according to claim 5, characterized in that, The supporting relationships upon which the supplementary results of the previous stage are based are the correspondence between the diagnostic basis items, examination basis items, and treatment path triggering basis items recorded in the supplementary results of the previous stage and the corresponding medication category, stage continuation status, and subsequent treatment path of the supplementary results of the previous stage. If the diagnostic information and examination results reached in the current stage do not negate the aforementioned basis, and there is no new basis in the current stage to rewrite the current supplementary candidate as another medication category, another stage of continued use status, or other subsequent treatment path, then the supporting relationship is determined to still be valid in the current stage. For supplementary candidates that meet the aforementioned conditions, the supplementary candidate identifier and supporting basis identifier corresponding to the supplementary results of the previous stage are used. For supplementary candidates that are not covered, a supplementary result for this round is generated that includes the identifier of the site to be supplemented, the supplementary candidate identifier, the medication category, the stage of continued use status, the subsequent treatment path, and the supporting basis identifier. The results used in the previous round are then merged with the supplementary results for this round to form the supplementary result for the current stage.

7. The medical expense insurance simulation decision-making method based on data fusion according to claim 1, characterized in that, For each site to be supplemented, receive subsequent diagnostic updates, examination updates, and processing path updates that arrive at the current stage, and identify the items associated with the corresponding site to be supplemented as the basis for new additions.

8. The medical expense insurance simulation decision-making method based on data fusion according to claim 7, characterized in that, The boundary item reopening mapping table records at least the identifier of the site to be filled, the identifier of the boundary item, the type of new basis that can trigger reopening, and the range of affected supplementary candidates. When the basis type, boundary item identifier, and the identifier of the site to be filled corresponding to the new basis are consistent with the records in the boundary item reopening mapping table, it is determined that the new basis hits the boundary item reopening mapping table corresponding to the site to be filled.

9. The medical expense insurance simulation decision-making method based on data fusion according to claim 8, characterized in that, When a new basis negates at least one of the diagnostic basis, examination basis, or treatment path triggering basis items upon which the previous supplementary result relied, or when the original basis combination no longer supports the medication category, stage continuation status, or subsequent treatment path corresponding to the original supplementary candidate, the supporting relationship is determined to be invalid. Supplementary results are regenerated only for the affected uncovered supplementary candidates in the sites to be supplemented where the boundary item is reopened and the supporting relationship is no longer valid. For supplementary candidates where the boundary item is not reopened or the supporting relationship is still valid, the original supplementary results are continued to be used. The regenerated supplementary results are merged with the original supplementary results that are continued to be used to obtain the updated supplementary results. The insurance simulation decision input is output, which includes the site to be supplemented identifier, supplementary candidate identifier, current effective medication category, current effective stage continuation status, current effective subsequent treatment path, and current effective supporting basis identifier.

10. A medical expense insurance simulation decision-making system based on data fusion, used to implement the medical expense insurance simulation decision-making method based on data fusion as described in any one of claims 1 to 9, characterized in that, It includes a site identification and candidate retention unit, a supplementary candidate merging and coverage determination unit, a new basis triggering local reopening and input update unit, and a rule relationship and determination caliber storage area; the site identification and candidate retention unit, the supplementary candidate merging and coverage determination unit, and the new basis triggering local reopening and input update unit are respectively used to execute S101, S102, and S103; the rule relationship and determination caliber storage area is used to store and be called by the above units the stage treatment plan description, medication rule relationship, correspondence between diagnostic examination and medication, supplementary candidate merging caliber, supporting relationship determination caliber, boundary item reopening registration caliber, new basis hit determination caliber, and new basis type caliber.

Images