Complaint text generation method, system, device and medium based on large language model

Abstract

The application provides a complaint text generation method, system, device and medium based on a large language model, comprising: obtaining medical data currently needing to be processed; processing the medical data currently needing to be processed to obtain corresponding text description information as current medical record text information; generating corresponding prompt words based on the current medical record text information and current reference text information corresponding to the medical data currently needing to be processed; inputting the current medical record text information and the prompt words into a large language model to obtain corresponding analysis results as a current processing result; verifying the current processing result to determine whether the current processing result is a usable result; and if it is determined that the current processing result is a usable result, generating corresponding complaint text and intermediate record text by using the large language model. The application can improve the generation efficiency and accuracy of complaint text.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of data processing, and in particular to a complaint text generation method, system, device and medium based on a large language model. BACKGROUND

[0002] In some application scenarios, given data needs to be analyzed to determine whether the set content associated with the given data meets the set condition. If the set condition is met, a corresponding complaint text will be formed. For example, in a medical scenario, complaint materials need to be submitted for medical insurance complaints. Currently, the given data is analyzed and the corresponding text content is formed by manual methods. This method has the problems of low efficiency and low accuracy. SUMMARY

[0003] To solve the above technical problems, the technical solution adopted by the present application is as follows:

[0004] According to the first aspect of the present application, a complaint text generation method based on a large language model is provided, which comprises the following steps:

[0005] S100, obtaining medical data currently needing to be processed, the medical data comprising case data and cost data associated with the case data.

[0006] S200, processing the medical data currently needing to be processed to obtain corresponding text description information as current medical record text information.

[0007] S300, generating corresponding prompt words based on the current medical record text information and current reference text information corresponding to the medical data currently needing to be processed; the current reference text information comprising at least one case description information allowing complaint.

[0008] S400, inputting the current medical record text information and the prompt words into a large language model to obtain a corresponding analysis result as a current processing result.

[0009] S500, verifying the current processing result to determine whether the current processing result is a usable result, and if the current processing result is determined to be a usable result, generating a corresponding complaint text and an intermediate record text using the large language model, the intermediate record text comprising process description information of the large language model obtaining the current processing result and description information representing that the similarity between the current medical record text information and at least one case description information in the current reference text information is greater than a set similarity threshold.

[0010] According to the second aspect of the present application, a complaint text generation system based on a large language model is provided, which comprises:

[0011] a data acquisition module configured to acquire medical data currently in need of processing, the medical data including case data and expense data associated with the case data.

[0012] a data processing module configured to process the medical data currently in need of processing to obtain corresponding text description information as current medical record text information.

[0013] a prompt word generation module configured to generate corresponding prompt words based on the current medical record text information and current reference text information corresponding to the medical data currently in need of processing, the current reference text information including at least one case description information that allows for filing a complaint.

[0014] an analysis module configured to input the current medical record text information and the prompt words into a large language model to obtain corresponding analysis results as current processing results.

[0015] a verification module configured to verify the current processing results to determine whether the current processing results are usable results, and if the current processing results are determined to be usable results, generate corresponding complaint text and intermediate record text using the large language model, the intermediate record text including process description information of the large language model obtaining the current processing results and description information representing that a similarity between the current medical record text information and at least one case description information in the current reference text information is greater than a set similarity threshold.

[0016] According to a fourth aspect of the present application, there is provided a computer readable storage medium storing programs or instructions for causing a computer to perform the steps of the method according to the first aspect of the present application.

[0017] The present application has at least the following beneficial effects:

[0018] The complaint text generation method based on the large language model provided by the embodiments of the present application can automatically generate complaint text and improve the generation efficiency and accuracy of complaint text due to the use of the large language model technology.

[0019] It should be understood that the content described in this section is not intended to identify key or important features of the embodiments of the present application, nor is it used to limit the scope of the present application. Other features of the present application will become apparent through the following description. BRIEF DESCRIPTION OF DRAWINGS

[0020] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings needed to be used in the embodiments will be briefly introduced. Obviously, the drawings in the following description only constitute some embodiments of the present application, and for those skilled in the art, other drawings can be obtained without creative effort based on these drawings.

[0021] Figure 1 The flow chart of the complaint text generation method based on a large language model provided by the embodiments of the present application is shown in

[0022] Figure 2 The structural block diagram of the complaint text generation system based on a large language model provided by the embodiments of the present application is shown in DETAILED DESCRIPTION

[0023] The technical solutions in the embodiments of the present application will be described clearly and completely with reference to the drawings of the embodiments of the present application. Obviously, the described embodiments only constitute some embodiments of the present application, rather than all the embodiments. Based on the embodiments of the present application, all other embodiments obtained by those skilled in the art without creative effort fall within the scope of the present application.

[0024] 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 the present application belongs. The terminology used in the description of the present application herein only for the purpose of describing the specific embodiments and is not intended to limit the present application. The term "and / or" as used herein includes any and all combinations of one or more of the associated listed items.

[0025] It should be noted that some exemplary embodiments are described as processes or methods that are depicted as flow diagrams. Although the processes are described in a particular sequential order, many of the steps can be performed in parallel, concurrently or simultaneously. In addition, the order of the steps can be re-arranged. The processes can be terminated when their operations are completed, but can also have additional steps not included in the figure, which can also be performed after the operations of the processes are completed. The processes can correspond to methods, functions, procedures, subroutines, subprograms, etc.

[0026] The embodiments of the present application provide a complaint text generation method based on a large language model, as shown in Figure 1 The method comprises the following steps:

[0027] S100, obtaining medical data currently to be processed, wherein the medical data comprises case data and cost data associated with the case data.

[0028] In the embodiments of the present application, the relevant case data and associated cost data can be obtained from the authorized electronic medical record system through an API interface or a database connection. The case data can include all the record information of the patient during the visit, such as diagnosis results, treatment plans, operation records, etc.

[0029] In the embodiments of the present application, the medical data currently in need of processing can be medical data that is identified by the relevant department as not meeting the preset regulations, for example, medical data that is identified by the medical insurance bureau as not meeting the medical insurance rules.

[0030] S200, processing the medical data currently in need of processing to obtain corresponding text description information as the current medical record text information.

[0031] In the embodiments of the present application, the case data and the cost data can be processed respectively to generate text descriptions about the case data and the cost data, for example, what operation was performed at a certain time, what drug was prescribed at a certain time, and how much money was charged, etc.

[0032] S300, generating a corresponding prompt word based on the current medical record text information and the current reference text information corresponding to the medical data currently in need of processing; the current reference text information includes at least one case description information that allows for appeal.

[0033] In a specific embodiment of the present application, the reference text information can be medical insurance rules, and the case description information can be a case condition for which special exceptions are reported, for example, the case description information can include: the number of inpatient days exceeds 5 times (including) the average inpatient days of the same level of the designated medical institution in the first half of the year for the disease group (DRG) / the disease (DIP) (each unified area can appropriately reduce the multiple according to the specific circumstances); the number of intensive care unit bed days exceeds 60% (including) of the total number of inpatient bed days for the case.

[0034] In the embodiments of the present application, the prompt word can be determined based on the actual situation.

[0035] In one illustrative embodiment, S300 can specifically include:

[0036] S301, obtaining the similarity between each case description information in the current medical record text information and the current reference text information to obtain at least one similarity.

[0037] S302, if there is a similarity greater than a set similarity threshold value in the at least one similarity, taking the similarity greater than the set similarity threshold value as a candidate similarity.

[0038] In the embodiments of the present application, the set similarity threshold value can be an empirical value.

[0039] The skilled in the art knows that if there is no similarity greater than the set similarity threshold in the at least one similarity, the corresponding prompt word can be generated directly based on the current medical record text information and the current reference text information corresponding to the current medical data to be processed.

[0040] In S303, a score corresponding to each candidate similarity is obtained, and a weighted sum of each candidate similarity and the corresponding score is obtained to obtain at least one weighted sum.

[0041] In an embodiment of the present application, the score corresponding to the candidate similarity can be determined based on the medical insurance reimbursement fee corresponding to the case description information corresponding to the candidate similarity, and specifically, the score can be positively correlated with the medical insurance reimbursement fee, that is, the greater the medical insurance reimbursement fee, the greater the corresponding score.

[0042] In an embodiment of the present application, the weight of the similarity and the weight of the score can be determined based on the actual situation, and in one illustrative embodiment, the weight of the similarity and the weight of the score can be the same, and in another illustrative embodiment, the weight of the similarity can be less than the weight of the score.

[0043] In S304, the case description information corresponding to the maximum weighted sum in the at least one weighted sum is taken as the information of interest, and a prompt word prompt information is generated based on the information of interest.

[0044] In an embodiment of the present application, the prompt word is used to guide the reasoning of the large language model. The prompt word prompt information can be determined based on the actual situation, for example, it can be "please prioritize matching the information of interest" and the like.

[0045] In S305, if an input instruction for generating a prompt word based on the prompt word prompt information is received, the prompt word is generated based on the current medical record text information, the current reference text information and the prompt word prompt information, and if an input instruction for generating a prompt word based on the prompt word prompt information is not received, the prompt word is generated based on the current medical record text information and the current reference text information.

[0046] In an embodiment of the present application, if the way of generating a prompt word based on the current medical record text information, the current reference text information and the prompt word prompt information, compared to the way of generating a prompt word based only on the current medical record text information and the current reference text information, can save processing time and make the matching result more accurate.

[0047] In an illustrative embodiment of the present application, one example of generating the prompt word based on the current medical record text information and the current reference text information can be as follows:

[0048]

[0049]

[0050] Example:

[0051] Suppose the following rules are included:

[0052] No. 16: The same-day charge code contains 003107010030000-310701003 exactly, and the number of charges on the same day > 1.

[0053] Rule 21: The same-day charge code matches 003315010520000-331501052 exactly, and the charge code matches 003302040010000-330204001 exactly.

[0054] Suppose the medical record contains the following content:

[0055] Patient Yang XX, female, age 27, admitted on November 17, 2024, discharged on November 23, 2024, actual hospitalization 6 days. Department of gynecology: medical insurance settlement level: three, diagnosis code and name include: N73.003: acute female pelvic inflammation, B37.301: Candida vaginitis, Z39.100: supervision during lactation, A09.02.07.03: pelvic inflammation, 003107010030000-310701003 was used on 2024-12-31, quantity 2 times. 003315010520000-331501052 was used on 2024-12-30.

[0056] Output the result in Chinese:

[0057]

[0058]

[0059] S400, input the current medical record text information and the prompt word into the large language model, get the corresponding analysis result as the current processing result.

[0060] In the embodiments of the application, the large language model can be an existing open source large language model.

[0061] S500, verify the current processing result to determine whether the current processing result is a usable result, if it is determined that the current processing result is a usable result, generate a corresponding complaint text and intermediate record text using the large language model.

[0062] In the embodiment of the present application, the intermediate record text comprises process description information of the current processing result obtained by the large language model and description information representing that the similarity between the current medical record text information and at least one case description information in the current reference text information is greater than a set similarity threshold.

[0063] As known by those skilled in the art, the large language model can automatically generate the intermediate record text based on the analysis result. The similarity can be a semantic similarity, a feature-based similarity, etc., and is preferably a semantic similarity.

[0064] Further, in an illustrative embodiment, in S500, the current processing result is verified to determine whether the current processing result is a usable result, specifically comprising:

[0065] S501, if the current processing result is a result representing that the similarity between the current medical record text information and at least one case description information in the current reference text information is greater than a set similarity threshold, it indicates that the expense reimbursement of the medical data to be processed is in line with the expense reimbursement rules of the current reference text information, and S502 is executed.

[0066] S502, determining an appeal expense reference value corresponding to the medical data to be processed based on the occurrence time of the medical data to be processed and the occurrence frequency of the expense-related object in the geographical area corresponding to the medical data to be processed.

[0067] In the embodiment of the present application, the geographical area can be an area divided according to an administrative unit. The expense-related object can be a disease recorded in the medical data that needs to be treated. The appeal expense reference value can be a medical insurance reimbursement expense that can be reimbursed.

[0068] As known by those skilled in the art, each case description information corresponds to an appeal expense reference value.

[0069] S503, comparing the expense information in the current processing result with the appeal expense reference value corresponding to the medical data to be processed, and if the comparison result represents that the absolute value of the difference between the expense information in the current processing result and the appeal expense reference value is less than a set value, it is determined that the current processing result is a usable result, otherwise, it is determined that the current processing result is not a usable result.

[0070] In the embodiment of the present application, the set value can be determined based on actual conditions, for example, it can be an empirical value. If the comparison result represents that the absolute value of the difference between the expense information in the current processing result and the appeal expense reference value is less than the set value, it indicates that the expense reimbursement in the medical data to be processed is in line with the reimbursement in the geographical area, which is a relatively reasonable reimbursement.

[0071] In the embodiments of the present application, the complaint expense reference value is determined based on a preset function expression, and the preset function expression is determined based on a product of the first expense influence factor and the first associated weight and a product of the second expense influence factor and the second associated weight, wherein the first expense influence factor is a time interval between the occurrence time of the medical data and the effective start time of the corresponding latest reference text information, and the second expense influence factor is the number of occurrences of the expense associated object corresponding to the medical data in the geographic area corresponding to the medical data.

[0072] In an illustrative embodiment, the preset function expression can be represented as: y=f(r1, r2), wherein y is the complaint expense reference value, r1 is the first expense influence factor, r2 is the second expense influence factor, and f() is the preset function expression. f() can be set based on actual needs. In an illustrative embodiment, f() can be a multiple linear function of r1 and r2, and preferably a linear function of r1 and r2, i.e., y=k1xr1+k2xr2+a. k1 is the first associated weight, k2 is the second associated weight, and a is a constant.

[0073] Further, in the embodiments of the present application, the first associated weight and the second associated weight in the preset function expression are obtained by the following steps:

[0074] S10, obtaining a set of historical medical data in a geographic area corresponding to the medical data currently to be processed, wherein the set of historical medical data includes m historical data, each historical data includes a corresponding complaint expense, a time interval between the occurrence time of the corresponding medical data and the effective start time of the corresponding latest reference text information, and the number of occurrences of the expense associated object corresponding to the medical data in the geographic area; and m>1.

[0075] In the embodiments of the present application, the complaint expense refers to the medical insurance reimbursement expense. Those skilled in the art know that each case description information corresponds to a complaint expense.

[0076] S11, obtaining a set of associated weight combinations C=(C1, C2, …, Cm-1) based on the set of historical medical data and the preset function expression, wherein C is the i-th associated weight combination in C, C=(C, C), C is the i-th first associated weight, C is the i-th second associated weight, and i takes a value from 1 to m-1. i m-1 i i ir1 ir2 ir1 ir2

[0077] ​​​​​​​​In the embodiment of the present application, each historical data can be expressed by a function expression, that is, can be expressed as an equation, so that the combination set of correlation weights can be obtained by solving m equation groups.

[0078] S12, a square with a length of L1 along the x-axis direction and a length of L2 along the y-axis direction is drawn in the rectangular coordinate system; wherein, L1=(C r1 max -C r1 min ), C r1 max is the maximum time interval correlation weight in C, C r1 min is the minimum time interval correlation weight in C, C r1 max = max((C ir1 ) i＝1……m-1 ), C r1 min = min((C ir1 ) i＝1……m-1 ), L2=(C r2 max -C r2 min ), C r2 max is the maximum occurrence number correlation weight in C, C r2 min is the minimum occurrence number correlation weight in C, C r2 max = max((C ir2 ) i＝1……m-1 ), C r2 min = min((C ir2 ) i＝1……m-1 ); the left upper corner coordinate of the square is (C r1 min , C r2 min ).

[0079] In the embodiment of the present application, the scale unit of the x-axis in the rectangular coordinate system can be day, and the scale unit of the y-axis can be times.

[0080] S13, min((C jr1 -C (j-1)r1 ) j＝2……m-1 is obtained as the length of the cell along the x-axis direction, and min((C jr2 -C (j-1)r2 ) j＝2……m-1 is obtained as the length of the cell along the y-axis direction; the value of j is 2 to m-1.

[0081] S14, dividing the square into n cells by using the cells.

[0082] In the embodiment of the present application, since L1 may not be an integer multiple of min((C jr1 -C (j-1)r1 ) j＝2……m-1 , L2 may not be an integer multiple of min((C jr2 -C (j-1)r2 ) j＝2……m-1 , the total area of the n cells can be greater than or equal to the area of the square.

[0083] S15, for the rth cell in the n cells, traversing C, if C ir1 belongs to the rth cell, adding C ir1 to the counting set corresponding to the rth cell, the initial value of the counting set corresponding to the rth cell is empty, and r takes values from 1 to n, and the initial value is 1.

[0084] S16, taking the cell corresponding to the counting set with the most associated weights in the n counting sets as the target cell.

[0085] In the embodiment of the present application, if the number of associated weights in a certain counting set is the most, it means that more associated weights are located in the cell corresponding to the counting set, which can be used as a representative associated weight.

[0086] S17, obtaining the first associated weight and the holiday associated weight in the preset function expression based on the target cell.

[0087] Further, in an illustrative embodiment, the first associated weight is the x coordinate value of the center coordinate of the target cell, and the second associated weight is the y coordinate value of the center coordinate of the target cell.

[0088] Further, in another illustrative embodiment, the first associated weight is the x coordinate value of the lower right corner coordinate of the target cell, and the second associated weight is the y coordinate value of the lower right corner coordinate of the target cell.

[0089] Further, in another embodiment of the present application, in S500, the verification of the current processing result to determine whether the current processing result is a usable result specifically includes:

[0090] S510, if the current processing result is a result representing that the similarity between at least one case description information in the current medical record text information and the current reference text information is greater than a set similarity threshold, performing S511.

[0091] S511, determining a complaint expense reference value corresponding to the medical data currently requiring processing based on the occurrence time of the medical data currently requiring processing and the occurrence times of the expense association object in the geographical area corresponding to the medical data currently requiring processing;

[0092] S512, comparing the expense information in the current processing result with the complaint expense reference value corresponding to the medical data currently requiring processing, and if the comparison result indicates that the absolute value of the difference between the expense information in the current processing result and the complaint expense reference value is less than a set value, determining that the current processing result is a usable result, otherwise, performing S513.

[0093] S513, determining whether the effective start time corresponding to the current reference text information is earlier than the effective end time of the last reference text information, and if the effective start time corresponding to the current reference text information is earlier than the effective end time of the last reference text information and the score of the analysis result corresponding to the current reference text information is greater than the score of the analysis result corresponding to the last reference text information, determining that the current processing result is a usable result.

[0094] In the embodiment of the application, the score of the analysis result is determined based on the medical insurance reimbursement expense corresponding to the case description information, and specifically, the score is positively correlated with the medical insurance reimbursement expense, that is, the greater the medical insurance reimbursement expense, the greater the corresponding score.

[0095] Compared with the foregoing embodiments, in the embodiment, when the comparison result indicates that the absolute value of the difference between the expense information in the current processing result and the complaint expense reference value is greater than or equal to the set value, whether the current processing result is a usable result is further determined based on the effective start time corresponding to the current reference text information and the effective end time of the last reference text information, so that the obtained processing result is more accurate, and the generation accuracy of the complaint text is improved.

[0096] Further, in the embodiment of the application, if the effective start time corresponding to the current reference text information is later than the effective end time of the last reference text information, it indicates that the used reference text information is accurate, and it is determined that the current processing result is not a usable result.

[0097] Further, in the embodiment of the application, if the effective start time corresponding to the current reference text information is earlier than the effective end time of the last reference text information and the score of the analysis result corresponding to the current reference text information is greater than the score of the analysis result corresponding to the last reference text information, the current processing result is determined to be a usable result, and specifically, the method can include:

[0098] If the effective start time corresponding to the current reference text information is earlier than the effective end time of the last reference text information, the large language model is instructed to analyze the current medical record text information and the last reference text information to obtain a corresponding current processing result as a reference result.

[0099] If the score of the analysis result corresponding to the current reference text information is greater than the score of the analysis result corresponding to the last reference text information, it is determined that the current processing result is a usable result.

[0100] Further, S513 further includes the following steps:

[0101] If the score corresponding to the current processing result is less than the score corresponding to the reference result, it is determined that the current processing result is a usable result, and the large language model is instructed to generate the complaint text and the intermediate record text based on the reference result.

[0102] In another embodiment of the present application, for target medical data identified by the relevant department as meeting the preset regulations, the following operations can be performed:

[0103] It is determined whether the effective start time corresponding to the reference text information used by the target medical data is earlier than the effective end time of the last reference text information. If the effective start time corresponding to the reference text information used is earlier than the effective end time of the last reference text information, and the score of the analysis result corresponding to the reference text information used is less than the score of the analysis result corresponding to the last reference text information, a corresponding complaint text is generated based on the analysis result corresponding to the last reference text information.

[0104] As known by those skilled in the art, if the effective start time corresponding to the reference text information used is earlier than the effective end time of the last reference text information, and the score of the analysis result corresponding to the reference text information used is greater than the score of the analysis result corresponding to the last reference text information, no processing is performed.

[0105] In the embodiments of the present application, since the above operations are performed for target medical data identified by the relevant department as meeting the preset regulations, the processing of medical data can be more accurate.

[0106] Based on the same inventive concept, the embodiments of the present application provide a large language model-based complaint text generation system, as shown in Figure 2 The system can include:

[0107] A data acquisition module 1 is configured to acquire current medical data to be processed, wherein the medical data includes case data and cost data associated with the case data.

[0108] Data processing module 2 is used to process the medical data that needs to be processed to obtain the corresponding text description information, which is used as the text information of the current medical record.

[0109] The prompt word generation module 3 is used to generate corresponding prompt words based on the current medical record text information and the current reference text information corresponding to the medical data that needs to be processed; the current reference text information includes at least one case description information that allows for appeals.

[0110] Analysis module 4 is used to input the current medical record text information and the prompt words into the large language model to obtain the corresponding analysis results, which are used as the current processing results.

[0111] The verification module 5 is used to verify the current processing result to determine whether the current processing result is a usable result. If the current processing result is determined to be a usable result, the corresponding appeal text and intermediate record text are generated using the large language model. The intermediate record text includes process description information of the large language model obtaining the current processing result and description information indicating that the similarity between at least one case description information in the current medical record text information and the current reference text information is greater than a set similarity threshold.

[0112] This system can be used to perform Figure 1 The method shown in the illustrated embodiment can be used as a reference for understanding the functions that each functional module of the system can achieve. Figure 1 The embodiments shown are described in detail below.

[0113] This invention also provides an electronic device, including: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, the instructions being configured to perform the method described in this invention.

[0114] This invention also provides a computer-readable storage medium storing computer-executable instructions for performing the methods described in this invention.

[0115] It should be understood that the various forms of processes shown above can be used to reorder, add, or delete steps. For example, the steps described in this invention can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution disclosed in this invention can be achieved, and this is not limited herein.

[0116] The above detailed description does not limit the scope of the application. Various modifications, combinations, sub-combinations and alternatives can be made to the detailed description. Any modification, equivalent replacement and improvement etc. made within the spirit and principle of the application shall be included in the scope of the application.

Claims

1. A method for generating a complaint text based on a large language model, the method comprising: The method comprises the following steps: S100, acquiring medical data currently in need of processing, the medical data comprising case data and cost data associated with the case data; S200, processing the medical data currently in need of processing to obtain corresponding text description information as current medical record text information; S300, generating a corresponding prompt word based on the current medical record text information and current reference text information corresponding to the medical data currently in need of processing, the current reference text information comprising at least one case description information allowing for complaint; S400, inputting the current medical record text information and the prompt word into a large language model to obtain a corresponding analysis result as a current processing result; S500, verifying the current processing result to determine whether the current processing result is a usable result, and if the current processing result is determined to be a usable result, generating a corresponding complaint text and intermediate record text using the large language model, the intermediate record text comprising process description information of the large language model obtaining the current processing result and description information representing that a similarity between the current medical record text information and at least one case description information in the current reference text information is greater than a set similarity threshold.

2. The method of claim 1, wherein, In S500, the verification of the current processing result to determine whether the current processing result is a usable result specifically comprises: S501, if the current processing result is a result representing that a similarity between the current medical record text information and at least one case description information in the current reference text information is greater than a set similarity threshold, performing S502; S502, determining a complaint cost reference value corresponding to the medical data currently in need of processing based on an occurrence time of the medical data currently in need of processing and a number of occurrences of a cost association object in a geographical area corresponding to the medical data currently in need of processing; S503, comparing cost information in the current processing result with the complaint cost reference value corresponding to the medical data currently in need of processing, and if a comparison result represents that an absolute value of a difference between the cost information in the current processing result and the complaint cost reference value is less than a set value, determining that the current processing result is a usable result, otherwise, determining that the current processing result is not a usable result.

3. The method of claim 1, wherein, In S500, the verification of the current processing result to determine whether the current processing result is a usable result specifically comprises: S510, if the current processing result is a result representing that a similarity between the current medical record text information and at least one case description information in the current reference text information is greater than a set similarity threshold, performing S511; S511, determining a complaint cost reference value corresponding to the medical data currently in need of processing based on an occurrence time of the medical data currently in need of processing and a number of occurrences of a cost association object in a geographical area corresponding to the medical data currently in need of processing; S512, compare the expense information in the current processing result with a complaint expense reference value corresponding to the medical data currently needing to be processed, and if the comparison result indicates that the absolute value of the difference between the expense information in the current processing result and the complaint expense reference value is less than a set value, it is determined that the current processing result is a usable result, otherwise, S513 is performed; S513, determine whether the effective start time corresponding to the current reference text information is earlier than the effective end time of the last reference text information, if the effective start time corresponding to the current reference text information is earlier than the effective end time of the last reference text information, and the score of the analysis result corresponding to the current reference text information is greater than the score of the analysis result corresponding to the last reference text information, it is determined that the current processing result is a usable result.

4. The method according to claim 2 or 3, characterized in that, The complaint expense reference value is determined based on a preset function expression, and the preset function expression is determined based on the product of a first expense influence factor and a first correlation weight and the product of a second expense influence factor and a second correlation weight, wherein the first expense influence factor is a time interval between the occurrence time of the medical data and the effective start time of the corresponding latest reference text information, and the second expense influence factor is the number of occurrences of the expense-related object corresponding to the medical data in the geographic area corresponding to the medical data. The first correlation weight and the second correlation weight in the preset function expression are obtained by the following steps: S10, obtain a set of historical medical data in a geographic area corresponding to the medical data currently needing to be processed, wherein the set of historical medical data includes m historical data, each historical data includes a corresponding complaint expense, a time interval between the occurrence time of the corresponding medical data and the effective start time of the corresponding latest reference text information, and the number of occurrences of the expense-related object corresponding to the medical data in the geographic area; m>1; S11, Based on the historical medical dataset and the preset function expression, obtain the associated weight combination set C = (C1, C2, ..., C...). i , ..., C m-1 ), C i Let C be the i-th association weight combination in C. i =(C ir1 C ir2 ), C ir1 Let C be the weight of the i-th first association. ir2 Let i be the i-th second association weight, where i ranges from 1 to m-1; S12, draw a square with a length of L1 along the x-axis direction and a length of L2 along the y-axis direction in the rectangular coordinate system; wherein, L1=(C r1 max -C r1 min ), C r1 max is the weight associated with the maximum time interval in C, C r1 min is the weight associated with the minimum time interval in C, C r1 max = max((C ir1 ) i＝1……m-1 ), C r1 min = min((C ir1 ) i＝1……m-1 ), L2=(C r2 max -C r2 min ), C r2 max is the weight associated with the maximum occurrence times in C, C r2 min is the weight associated with the minimum occurrence times in C, C r2 max = max((C ir2 ) i＝1……m-1 ), C r2 min = min((C ir2 ) i＝1……m-1 ); the upper left corner coordinates of the square are (C r1 min , C r2 min ); S13, get min((C jr1 - C (j-1)r1 )) j＝2……m-1 as the length of the cell in the x-axis direction and get min((C jr2 - C (j-1)r2 )) j＝2……m-1 as the length of the cell in the y-axis direction; j takes values from 2 to m-1; S14, divide the square into n cells using the cells; S15, for the rth cell in n cells, traverse C, if C ir1 belongs to the rth cell, add C ir1 to the counting set corresponding to the rth cell, the initial value of the counting set corresponding to the rth cell is empty, r takes values from 1 to n, and the initial value is 1; S16, take the cell corresponding to the count set with the most combined correlation weights in the n count sets as a target cell; S17, obtain the first correlation weight and the second correlation weight in the preset function expression based on the target cell.

5. The method of claim 4, wherein, The first correlation weight is the x-coordinate value of the center coordinate of the target cell, and the second correlation weight is the y-coordinate value of the center coordinate of the target cell.

6. The method of claim 4, wherein, The first correlation weight is the x-coordinate value of the lower right corner coordinate of the target cell, and the second correlation weight is the y-coordinate value of the lower right corner coordinate of the target cell.

7. The method of claim 1, wherein, S300 specifically includes: S301, obtain the similarity between each case description information in the current medical record text information and the current reference text information to obtain at least one similarity; S302, if there is a similarity greater than a set similarity threshold in the at least one similarity, take the similarity greater than the set similarity threshold as a candidate similarity; S303, obtain the score corresponding to each candidate similarity, and obtain the weighted sum of each candidate similarity and the corresponding score to obtain at least one weighted sum; S304, taking the case description information corresponding to the maximum weighting sum in the at least one weighting sum as the information of interest, and generating prompt word prompt information based on the information of interest; S305, if an input instruction of generating a prompt word based on the prompt word prompt information is received, generating the prompt word based on the current medical record text information, the current reference text information, and the prompt word prompt information, and if an input instruction of generating a prompt word based on the prompt word prompt information is not received, generating the prompt word based on the current medical record text information and the current reference text information. 8.A system for generating a complaint text based on a large language model, characterized by, The system comprises: A data acquisition module is configured to acquire current medical data to be processed, wherein the medical data comprises case data and cost data associated with the case data. A data processing module is configured to process the current medical data to be processed to obtain corresponding text description information as current medical record text information. A prompt word generation module is configured to generate a corresponding prompt word based on the current medical record text information and current reference text information corresponding to the current medical data to be processed, wherein the current reference text information comprises at least one case description information that allows for complaint. An analysis module is configured to input the current medical record text information and the prompt word into a large language model to obtain a corresponding analysis result as a current processing result. A verification module is configured to verify the current processing result to determine whether the current processing result is a usable result, and if the current processing result is determined to be a usable result, generate a corresponding complaint text and an intermediate record text using the large language model, wherein the intermediate record text comprises process description information of the large language model obtaining the current processing result and description information representing that the similarity between the current medical record text information and at least one case description information in the current reference text information is greater than a set similarity threshold.

9. An electronic device, comprising: A processor and a memory are included. The processor is configured to execute the steps of the method according to any one of claims 1 to 7 by calling programs or instructions stored in the memory.

10. A computer-readable storage medium, characterized in that, The computer readable storage medium is configured to store programs or instructions, which enable the computer to execute the steps of the method according to any one of claims 1 to 7.

Images