Medical claim settlement auditing method and device, equipment and medium
Through the automated audit of preset medical audit models, the problems of time-consuming, labor-intensive and cost-effective manual audits are solved, and efficient and accurate medical claims review is achieved, improving user experience and reducing costs.
Patent Information
- Application Number
- CN202411127038.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-16
- Publication Date
- 2024-12-06
Smart Images

Figure CN119090653A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of insurance claims technology, and in particular to a medical claims review method, device, equipment and medium. Background Art
[0002] Medical claims review in the current market is conducted by hiring senior medical personnel for medical review. These professionals need to review the information provided by users and the diseases and drugs for which claims are applied, and need to know the current medical insurance policies, the terms of insurance products, etc., to conduct a comprehensive review of the rationality of medication. However, for specific terms, such as whether it is the first diagnosis, whether this claim is the first use of medication for first-line treatment, whether the treatment has been carried out according to the treatment plan specified in the product terms, etc., in addition to being familiar with the current case and product terms information, the reviewers may also need to review a lot of historical case information.
[0003] The above-mentioned manual review method has the following problems: (1) In terms of review time, it is necessary to repeatedly review and check the information, which is time-consuming and labor-intensive; (2) In terms of personnel, the reviewers are required to be familiar with insurance terms and medical knowledge, which places high demands on the quality of personnel; (3) For insurance companies, since the quality requirements for reviewers are high, and diseases and drugs are fixed, most of the work is repetitive, which generates a lot of labor costs; (4) For users, since manual review is time-consuming and labor-intensive, it takes a long time to wait for the results, which affects the customer experience. Summary of the invention
[0004] The present invention provides a medical claims review method, device, equipment and medium, which perform automated medical claims review through a preset medical review model, and can improve review efficiency, reduce labor costs, shorten claims timeliness and enhance user experience on the basis of ensuring review accuracy.
[0005] According to one aspect of the present invention, a medical claim review method is provided, the method comprising:
[0006] Obtaining pending medical claims information, and determining pending diseases, pending drugs, pending drug use indicator information, and current user profile information based on the pending medical claims information;
[0007] Inputting the diseases to be reviewed, the drugs to be reviewed and the drug use index information to be reviewed into a preset basic medical model to obtain medical proper nouns to be reviewed; wherein the preset basic medical model is used to parse medical proper nouns;
[0008] Determine historical user portrait information corresponding to the current user portrait information based on the disease to be reviewed, and determine the user portrait information to be reviewed based on the current user portrait information and the historical user portrait information;
[0009] Determine the reference drug indications corresponding to the drug to be reviewed, input the disease to be reviewed, the drug to be reviewed, the medication indicator information to be reviewed, the medical term to be reviewed, the user portrait information to be reviewed, and the reference drug indications into a preset medical review model to obtain a medical review result.
[0010] According to another aspect of the present invention, a medical claim review device is provided, comprising:
[0011] A pending medical claims information processing module is used to obtain pending medical claims information and determine pending diseases, pending drugs, pending drug use index information and current user portrait information based on the pending medical claims information;
[0012] A module for determining medical terms to be reviewed, which is used to input the diseases to be reviewed, the drugs to be reviewed and the drug use index information to be reviewed into a preset basic medical model to obtain medical terms to be reviewed; wherein the preset basic medical model is used to parse medical terms;
[0013] A module for determining user portrait information to be reviewed, used to determine historical user portrait information corresponding to the current user portrait information based on the disease to be reviewed, and to determine the user portrait information to be reviewed based on the current user portrait information and the historical user portrait information;
[0014] The medical review result determination module is used to determine the reference drug indications corresponding to the drug to be reviewed, and input the disease to be reviewed, the drug to be reviewed, the medication indicator information to be reviewed, the medical term to be reviewed, the user portrait information to be reviewed, and the reference drug indications into a preset medical review model to obtain the medical review result.
[0015] According to another aspect of the present invention, an electronic device is provided, the electronic device comprising:
[0016] at least one processor; and
[0017] a memory communicatively connected to the at least one processor; wherein,
[0018] The memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute the medical claim review method described in any embodiment of the present invention.
[0019] According to another aspect of the present invention, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a processor to implement the medical claims review method described in any embodiment of the present invention when executed.
[0020] The technical solution of the embodiment of the present invention first obtains the medical claim information to be reviewed, and determines the disease to be reviewed, the drug to be reviewed, the drug use index information to be reviewed, and the current user portrait information according to the medical claim information to be reviewed; then the disease to be reviewed, the drug to be reviewed, and the drug use index information to be reviewed are input into the preset basic medical model to obtain the medical proper nouns to be reviewed; wherein the preset basic medical model is used to parse the medical proper nouns; then the historical user portrait information corresponding to the current user portrait information is determined based on the disease to be reviewed, and the user portrait information to be reviewed is determined according to the current user portrait information and the historical user portrait information; then the reference drug indication corresponding to the drug to be reviewed is determined, and the disease to be reviewed, the drug to be reviewed, the drug use index information to be reviewed, the medical proper nouns to be reviewed, the user portrait information to be reviewed, and the reference drug indication are input into the preset medical review model to obtain the medical review result. This technical solution performs automated medical claim review through the preset medical review model, which can improve the review efficiency, reduce labor costs, shorten the claim settlement time, and enhance the user experience on the basis of ensuring the accuracy of the review.
[0021] It should be understood that the contents described in this section are not intended to identify the key or important features of the embodiments of the present invention, nor are they intended to limit the scope of the present invention. Other features of the present invention will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS
[0022] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.
[0023] Figure 1 is a flow chart of a medical claims review method provided according to Embodiment 1 of the present invention;
[0024] Figure 2 is a flow chart of a medical claims review method provided according to Embodiment 2 of the present invention;
[0025] Figure 3 is a schematic diagram of the structure of a medical claim review device provided according to Embodiment 3 of the present invention;
[0026] Figure 4It is a structural schematic diagram of an electronic device for implementing a medical claims review method according to an embodiment of the present invention. DETAILED DESCRIPTION
[0027] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention.
[0028] It should be noted that the terms "first", "second", "target", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.
[0029] Embodiment 1
[0030] Figure 1 This is a flow chart of a medical claim review method provided in the first embodiment of the present invention. This embodiment is applicable to the situation of efficiently and accurately reviewing medical claim cases. The method can be executed by a medical claim review device. The medical claim review device can be implemented in the form of hardware and / or software. The medical claim review device can be configured in an electronic device with data processing capabilities. Figure 1 As shown, the method includes:
[0031] S110, obtaining the pending medical claims information, and determining the pending diseases, pending drugs, pending drug use index information and current user portrait information based on the pending medical claims information.
[0032] Among them, the medical claim information to be reviewed may refer to the medical claim information awaiting review, which can be specifically represented in the form of an image. Exemplarily, the medical claim information to be reviewed includes the policy information to be reviewed and the medical treatment information to be reviewed. Among them, the policy information to be reviewed may refer to the policy information awaiting review, and the medical treatment information to be reviewed may include diagnostic information (such as CT, ultrasound, MRI, etc.) and treatment information (such as prescriptions, invoices, etc.). The diseases to be reviewed and the drugs to be reviewed can be understood as the names of diseases and drugs awaiting review, respectively. The medication indicator information to be reviewed can be understood as the medication condition information awaiting review, such as age, time, number of times, applicable diseases, etc. Among them, the medication indicators can be pre-set by the auditor based on the audit experience. The current user portrait information may refer to the user portrait information corresponding to this medical claim application. Exemplarily, the user portrait information may include disease course information, treatment information, and medication information.
[0033] In this embodiment, the medical claim information to be reviewed is first obtained in the form of an image, and then the text information is parsed out through OCR (Optical Character Recognition) technology, and then the text information is handed over to a general large language model, and the large language model is used to extract the diseases to be reviewed, the drugs to be reviewed, the drug use index information to be reviewed, and the current user portrait information from the medical claim information to be reviewed. Optionally, the diseases to be reviewed, the drugs to be reviewed, the drug use index information to be reviewed, and the current user portrait information are determined based on the medical claim information to be reviewed, including: extracting the diseases to be reviewed and the drugs to be reviewed from the insurance policy information to be reviewed; extracting the drug use index information to be reviewed and the current user portrait information from the diagnosis and treatment information to be reviewed.
[0034] S120, inputting the information of the diseases to be reviewed, the drugs to be reviewed and the drug use indexes to be reviewed into the preset basic medical model to obtain the medical terms to be reviewed.
[0035] The preset basic medical model is used to parse medical terms. Exemplarily, the training process of the preset basic medical model is as follows: determining a training data set based on the collected medical knowledge data; training a preset network model based on the training data set, and determining the preset network model as the preset basic medical model when the preset network model converges.
[0036] Specifically, medical knowledge data can be collected through various channels, such as medical personnel, websites of the Medical Insurance Bureau and the Drug Administration, and historical claims data assets of insurance companies, and medical knowledge data can be extracted from them, such as what situations or indicators are HR2 positive, what is O drug, what combination drugs are R-CHOP, etc. Then, the preset network model is trained based on the training data set determined according to the medical knowledge data, and whether the model converges during the training process is determined. When the model reaches convergence, the model training process can be stopped, and the preset network model trained for the last time is determined as the preset basic medical model. Among them, the preset network model can refer to a pre-set neural network model, such as a deep learning network. After the training of the preset basic medical model is completed, the information of the diseases to be reviewed, the drugs to be reviewed, and the drug use indicators to be reviewed can be input into the preset basic medical model, and the medical terms to be reviewed can be obtained according to the model output.
[0037] S130, determining historical user portrait information corresponding to the current user portrait information based on the disease to be reviewed, and determining the user portrait information to be reviewed based on the current user portrait information and the historical user portrait information.
[0038] The historical user portrait information is used to describe the user portrait information of the historical candidate objects under the historical candidate diseases. The historical candidate objects can be used to characterize the users to whom the historical user portrait information belongs, and the historical candidate diseases can be used to characterize the diseases described by the historical user portrait information. Exemplarily, the historical user portrait information may include historical candidate objects, historical candidate diseases, medical history, treatment history, medication history, claim application history, and historical review conclusions.
[0039] In this embodiment, optionally, the method also includes: determining historical user portrait information based on historical medical claims information, and storing the historical user portrait information in a first database; correspondingly, determining historical user portrait information corresponding to the current user portrait information based on the disease to be reviewed, including: determining the historical candidate object matching the current user portrait information in the first database as the target object; determining the historical candidate disease matching the disease to be reviewed in the first database as the target disease; and obtaining the user portrait information under the target object and the target disease from the first database as the historical user portrait information corresponding to the current user portrait information.
[0040] Specifically, the general large language model can be used to extract information such as medical history, treatment history, medication history, claim application history, and historical review conclusions from the historical data assets of the insurance company (historical medical claims information) based on the user-disease-time dimension, and the extracted information is stored as historical user portrait information in the first database. For example, the first database can be a relational database.
[0041] After obtaining the disease to be reviewed and the current user portrait information, a historical candidate object matching the user object to which the current user portrait information belongs can be searched from the first database as the target object, and the historical candidate disease matching the disease to be reviewed can be searched from the first database as the target disease. Then, the user portrait information under the target object and the target disease is obtained from the first database and determined as the historical user portrait information corresponding to the current user portrait information. After determining the historical user portrait information corresponding to the current user portrait information, the current user portrait information and its corresponding historical user portrait information can be integrated to obtain the user portrait information to be reviewed.
[0042] S140, determine the reference drug indications corresponding to the drug to be reviewed, input the disease to be reviewed, the drug to be reviewed, the medication indicator information to be reviewed, the medical terminology to be reviewed, the user portrait information to be reviewed, and the reference drug indications into the preset medical review model to obtain the medical review results.
[0043] Among them, the preset medical review model may refer to a network model that can realize medical claims review. The reference drug indication may refer to the drug indication that the drug to be reviewed should comply with. In this embodiment, optionally, the method also includes: determining the drug indication corresponding to each candidate drug according to the candidate insurance terms, and storing the candidate drugs and the drug indications corresponding to the candidate drugs in a second database; accordingly, determining the reference drug indication corresponding to the drug to be reviewed, including: determining the candidate drug in the second database that matches the drug to be reviewed as the target drug; obtaining the drug indication corresponding to the target drug from the second database as the reference drug indication corresponding to the drug to be reviewed.
[0044] Among them, the candidate insurance clauses may refer to the insurance clauses provided by the insurance company, such as the insurance clauses of Huiminbao. The candidate drugs may refer to the drugs involved in the candidate insurance clauses. Specifically, after the insurance product is launched, the insurance clauses have been determined. Especially for Huiminbao, the difference between the products of each period is not big. The insurance clauses corresponding to different types of insurance products can be used as candidate insurance clauses. For each candidate insurance clause, the drug indication elements corresponding to each candidate drug in the candidate insurance clause can be disassembled through RAG (Retrieval Augmented Generation) and LLM (Large Language Model). In order to ensure the accuracy of the drug indication elements, it is necessary to verify and correct the data by medical personnel, and then store the corrected data (candidate drugs and their corresponding drug indications) in the second database. Exemplarily, the second database can be a relational database. In the above process, after the model analysis and disassembly, only one medical personnel is required to review and correct the drug indication elements, and a large number of medical personnel are not required to disassemble the drug indication elements, thereby greatly reducing the labor cost.
[0045] When determining the reference drug indications corresponding to the drug to be reviewed, candidate drugs matching the drug to be reviewed can be searched from the second database as target drugs, and the drug indications corresponding to the target drug can be obtained from the second database as reference drug indications corresponding to the drug to be reviewed. After determining the reference drug indications corresponding to the drug to be reviewed, the disease to be reviewed, the drug to be reviewed, the drug to be reviewed, the drug use indicator information to be reviewed, the medical term to be reviewed, the user portrait information to be reviewed, and the reference drug indications can be input into the preset medical review model, and the medical review results are given through model analysis and reasoning. Among them, the medical review results include review passed or review failed. Specifically, if the review is passed, the reason for passing needs to be given, which reference drug indication is met, what drug use indicators are required for this reference drug indication, and what is extracted from the user portrait information to be reviewed; if the review is not passed, it is necessary to explain which ones do not meet the reference drug indications.
[0046] The technical solution of the embodiment of the present invention first obtains the medical claim information to be reviewed, and determines the disease to be reviewed, the drug to be reviewed, the drug use index information to be reviewed, and the current user portrait information according to the medical claim information to be reviewed; then the disease to be reviewed, the drug to be reviewed, and the drug use index information to be reviewed are input into the preset basic medical model to obtain the medical proper nouns to be reviewed; wherein the preset basic medical model is used to parse the medical proper nouns; then the historical user portrait information corresponding to the current user portrait information is determined based on the disease to be reviewed, and the user portrait information to be reviewed is determined according to the current user portrait information and the historical user portrait information; then the reference drug indication corresponding to the drug to be reviewed is determined, and the disease to be reviewed, the drug to be reviewed, the drug use index information to be reviewed, the medical proper nouns to be reviewed, the user portrait information to be reviewed, and the reference drug indication are input into the preset medical review model to obtain the medical review result. This technical solution performs automated medical claim review through the preset medical review model, which can improve the review efficiency, reduce labor costs, shorten the claim settlement time, and enhance the user experience on the basis of ensuring the accuracy of the review. For auditors, the workload is greatly reduced. The work of reviewing case-by-case documents, insurance clauses, and whether the indication information matches has been transformed into AI-based case sample spot checks and compounding of uncertain conclusions. For insurance companies, they can do more with less manpower. In the past, a large number of auditors with professional knowledge were needed, but now only a small number of people are needed to maintain basic data and review a small number of audit cases. For users, after the application information is uploaded, the personnel are skipped directly and the model performs automated review. Compared with the past, this greatly shortens the waiting time, improves the user experience, and reduces the burden on users in the shortest time.
[0047] In this embodiment, optionally, before extracting the pending diseases and pending drugs from the pending policy information, the method also includes: performing policy verification based on the pending policy information to obtain a verification result; wherein the verification result is pass or fail; and determining whether to initiate medical claims review based on the verification result.
[0048] It should be noted that before the actual medical claim review is conducted, there may be a situation where the policy is not approved, such as the insured's conditions do not meet the requirements, the policy loan conditions are insufficient, etc. At this time, in order to save review resources and time costs, the medical claim review can be stopped in time. In this embodiment, before extracting the pending diseases and pending drugs from the pending policy information, a policy verification link is added. Specifically, first, the policy is verified based on the pending policy information to obtain a verification result (pass or fail), and then it is determined whether to start the medical claim review based on the verification result. If the verification passes, the subsequent medical claim review process is started; if the verification fails, the subsequent medical claim review process is not started, and the policy verification result is fed back to the interactive interface to remind the user that the policy has not passed.
[0049] Through such a setting, this plan can avoid starting the subsequent medical claim review process if the policy verification fails, which can effectively save review resources and time costs.
[0050] In this embodiment, optionally, the training process of the preset medical review model includes: obtaining historical manual review data, and performing data cleaning on the historical manual review data to obtain training data; performing model training based on the training data to obtain a preset medical review model, and performing performance verification on the preset medical review model based on the model review results of the preset medical review model and the manual review results of the historical manual review data; if the preset medical review model fails the performance verification, training the preset basic medical model based on the training data, and tuning the preset medical review model based on the training results of the preset basic medical model until the preset medical review model passes the performance verification.
[0051] Specifically, when training the preset medical review model, first obtain the historical manual review data, clean the dirty data in the historical manual review data to obtain the training data, and then use the cleaned training data for the training of the preset medical review model. After the training of the preset medical review model is completed, in order to verify the effectiveness of the model, the model needs to be performance verified. Specifically, firstly, the performance of the preset medical review model is verified according to the model review result of the preset medical review model and the manual review result of the historical manual review data. Among them, if the model review result is consistent with the manual review result, it indicates that the model is valid. At this time, it is determined that the preset medical review model has passed the performance verification, and the model can be used for medical claims review; if the model review result is inconsistent with the manual review result, it indicates that the model is invalid. At this time, it is determined that the preset medical review model has not passed the performance verification, and the model cannot be used for medical claims review. Further, if the preset medical review model has not passed the performance verification, the preset basic medical model can be trained according to the training data, and the preset medical review model can be tuned based on the training result of the preset basic medical model until the preset medical review model passes the performance verification.
[0052] Through such a setting, this scheme can ensure the effectiveness of the model by verifying the performance of the preset medical review model, which helps to improve the accuracy of medical claims review.
[0053] Embodiment 2
[0054] Figure 2 A flowchart of a medical claims review method provided in Embodiment 2 of the present invention, this embodiment is optimized based on the above embodiment. The specific optimization is: the method also includes: obtaining medical proper nouns as reference medical proper nouns, and storing the reference medical proper nouns in a third database; correspondingly, after obtaining the medical proper nouns to be reviewed, the method also includes: based on the reference medical proper nouns stored in the third database, normalizing the data of the diseases to be reviewed, the drugs to be reviewed, the drug use index information to be reviewed, the current user portrait information, and the medical proper nouns to be reviewed.
[0055] like Figure 2 As shown, the method of this embodiment specifically includes the following steps:
[0056] S210, obtaining the pending medical claims information, and determining the pending diseases, pending drugs, pending drug use index information and current user portrait information based on the pending medical claims information.
[0057] S220, input the information of the diseases to be reviewed, the drugs to be reviewed, and the drug use indexes to be reviewed into the preset basic medical model to obtain the medical terms to be reviewed.
[0058] The preset basic medical model is used to parse medical terms. The specific implementation of S210-S220 can refer to the detailed description of S110-S120, which will not be repeated here.
[0059] S230, based on the reference medical terms stored in the third database, normalize the data of the diseases to be reviewed, the drugs to be reviewed, the drug use index information to be reviewed, the current user portrait information and the medical terms to be reviewed.
[0060] It should be noted that in the process of information identification and extraction, there may be inaccurate information identification (such as incomplete or wrong) or diversified information expression, which may affect the medical review results. Therefore, in order to further improve the accuracy of medical review, it is necessary to perform data normalization on the data input into the preset medical review model based on the reference medical terms stored in the third database, so that the incorrectly identified medical terms are converted into correct medical terms, and different expressions of the same parameter are converted into the same expression, thereby avoiding medical review errors due to inaccurate information identification or information diversity. Among them, the reference medical terms can be obtained and embedded and stored at the same time when the medical knowledge data required for the training of the preset basic medical model is collected. Exemplarily, the third database can be a vector database; the reference medical terms can include drug names, generic drug names (chemical names), disease names, medical indicator names, etc.
[0061] S240, determining historical user portrait information corresponding to the current user portrait information based on the disease to be reviewed, and determining the user portrait information to be reviewed based on the current user portrait information and the historical user portrait information.
[0062] S250, determine the reference drug indications corresponding to the drug to be reviewed, input the disease to be reviewed, the drug to be reviewed, the medication indicator information to be reviewed, the medical terminology to be reviewed, the user portrait information to be reviewed, and the reference drug indications into the preset medical review model to obtain the medical review results.
[0063] It should be noted that the information involved in S240-S250 is all information after data normalization. The specific implementation method can refer to the detailed description in S130-S140, which will not be repeated here.
[0064] The technical solution of the embodiment of the present invention, after obtaining the medical proper nouns to be reviewed, normalizes the data of the diseases to be reviewed, the drugs to be reviewed, the drug use index information to be reviewed, the current user portrait information and the medical proper nouns to be reviewed based on the reference medical proper nouns stored in the third database, and then determines the input data of the preset medical review model based on the information after the data normalization to perform automated medical claims review. This technical solution can further improve the review accuracy and enhance the user experience on the basis of improving the review efficiency, reducing labor costs and shortening the claims time limit.
[0065] Embodiment 3
[0066] Figure 3 This is a schematic diagram of the structure of a medical claim review device provided in the third embodiment of the present invention. The device can execute the medical claim review method provided in any embodiment of the present invention and has the corresponding functional modules and beneficial effects of the execution method. Figure 3 As shown, the device comprises:
[0067] The pending medical claim information processing module 310 is used to obtain pending medical claim information and determine pending diseases, pending drugs, pending drug use index information and current user portrait information according to the pending medical claim information;
[0068] The module 320 for determining the medical terms to be reviewed is used to input the diseases to be reviewed, the drugs to be reviewed and the drug use index information to be reviewed into a preset basic medical model to obtain the medical terms to be reviewed; wherein the preset basic medical model is used to parse the medical terms;
[0069] A module 330 for determining user portrait information to be reviewed, configured to determine historical user portrait information corresponding to the current user portrait information based on the disease to be reviewed, and to determine the user portrait information to be reviewed based on the current user portrait information and the historical user portrait information;
[0070] The medical review result determination module 340 is used to determine the reference drug indications corresponding to the drug to be reviewed, and input the disease to be reviewed, the drug to be reviewed, the medication indicator information to be reviewed, the medical term to be reviewed, the user portrait information to be reviewed, and the reference drug indications into a preset medical review model to obtain the medical review result.
[0071] Optionally, the medical claim information to be reviewed includes insurance policy information to be reviewed and medical treatment information to be reviewed;
[0072] Accordingly, the pending medical claims information processing module 310 is used to:
[0073] Extracting the pending disease and the pending drug from the pending policy information;
[0074] The medication index information to be reviewed and the current user portrait information are extracted from the medical information to be reviewed.
[0075] Optionally, the pending medical claims information processing module 310 is further used to:
[0076] Before extracting the to-be-reviewed diseases and the to-be-reviewed drugs from the to-be-reviewed insurance policy information, performing insurance policy verification according to the to-be-reviewed insurance policy information to obtain a verification result; wherein the verification result is pass or fail;
[0077] Determine whether to initiate medical claims review based on the verification results.
[0078] Optionally, the device further comprises:
[0079] A historical user portrait information determination module, used to determine historical user portrait information according to historical medical claims information, and store the historical user portrait information in a first database; wherein the historical user portrait information is used to describe user portrait information of historical candidate objects under historical candidate diseases;
[0080] Accordingly, the module 330 for determining the user portrait information to be reviewed is used to:
[0081] Determine the historical candidate object in the first database that matches the current user portrait information as the target object;
[0082] Determine the historical candidate disease in the first database that matches the disease to be reviewed as a target disease;
[0083] The user portrait information under the target object and the target disease is obtained from the first database as the historical user portrait information corresponding to the current user portrait information.
[0084] Optionally, the device further comprises:
[0085] A drug indication determination module, used to determine the drug indication corresponding to each candidate drug according to the candidate insurance terms, and store the candidate drug and the drug indication corresponding to the candidate drug in a second database;
[0086] Accordingly, the medical review result determination module 340 is used to:
[0087] Determine the candidate drug in the second database that matches the drug to be reviewed as the target drug;
[0088] The drug indication corresponding to the target drug is obtained from the second database as the reference drug indication corresponding to the drug to be reviewed.
[0089] Optionally, the device further comprises:
[0090] A reference medical proper noun determination module, used to obtain a medical proper noun as a reference medical proper noun, and store the reference medical proper noun in a third database;
[0091] Accordingly, the device further comprises:
[0092] The data normalization module is used to normalize the data of the to-be-reviewed diseases, the to-be-reviewed drugs, the to-be-reviewed medication indicator information, the current user portrait information and the to-be-reviewed medical terms based on the reference medical terms stored in the third database after obtaining the to-be-reviewed medical terms.
[0093] Optionally, the device further comprises: a preset medical review model training module, which is used to:
[0094] Acquire historical manual review data, and perform data cleaning on the historical manual review data to obtain training data;
[0095] Performing model training based on the training data to obtain a preset medical review model, and performing performance verification on the preset medical review model according to a model review result of the preset medical review model and a manual review result of the historical manual review data;
[0096] If the preset medical review model fails the performance verification, the preset basic medical model is trained according to the training data, and the preset medical review model is tuned based on the training results of the preset basic medical model until the preset medical review model passes the performance verification.
[0097] A medical claims review device provided in an embodiment of the present invention can execute a medical claims review method provided in any embodiment of the present invention, and has functional modules and beneficial effects corresponding to the execution method.
[0098] Embodiment 4
[0099] Figure 4 A schematic diagram of the structure of an electronic device 10 that can be used to implement an embodiment of the present invention is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital processing, cellular phones, smart phones, wearable devices (such as helmets, glasses, watches, etc.) and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present invention described and / or required herein.
[0100] like Figure 4As shown, the electronic device 10 includes at least one processor 11, and a memory connected to the at least one processor 11, such as a read-only memory (ROM) 12, a random access memory (RAM) 13, etc., wherein the memory stores a computer program that can be executed by at least one processor, and the processor 11 can perform various appropriate actions and processes according to the computer program stored in the read-only memory (ROM) 12 or the computer program loaded from the storage unit 18 to the random access memory (RAM) 13. In the RAM 13, various programs and data required for the operation of the electronic device 10 can also be stored. The processor 11, the ROM 12, and the RAM 13 are connected to each other through a bus 14. An input / output (I / O) interface 15 is also connected to the bus 14.
[0101] A number of components in the electronic device 10 are connected to the I / O interface 15, including: an input unit 16, such as a keyboard, a mouse, etc.; an output unit 17, such as various types of displays, speakers, etc.; a storage unit 18, such as a disk, an optical disk, etc.; and a communication unit 19, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 19 allows the electronic device 10 to exchange information / data with other devices through a computer network such as the Internet and / or various telecommunication networks.
[0102] The processor 11 may be a variety of general and / or special processing components with processing and computing capabilities. Some examples of the processor 11 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, a digital signal processor (DSP), and any appropriate processor, controller, microcontroller, etc. The processor 11 executes the various methods and processes described above, such as a medical claims review method.
[0103] In some embodiments, the medical claims review method may be implemented as a computer program, which is tangibly contained in a computer-readable storage medium, such as a storage unit 18. In some embodiments, part or all of the computer program may be loaded and / or installed on the electronic device 10 via the ROM 12 and / or the communication unit 19. When the computer program is loaded into the RAM 13 and executed by the processor 11, one or more steps of the medical claims review method described above may be performed. Alternatively, in other embodiments, the processor 11 may be configured to execute the medical claims review method in any other appropriate manner (e.g., by means of firmware).
[0104] Various implementations of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), load programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various implementations can include: being implemented in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which can be a special purpose or general purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.
[0105] Computer programs for implementing the methods of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, so that when the computer program is executed by the processor, the functions / operations specified in the flow chart and / or block diagram are implemented. The computer program may be executed entirely on the machine, partially on the machine, partially on the machine and partially on a remote machine as a stand-alone software package, or entirely on a remote machine or server.
[0106] In the context of the present invention, a computer-readable storage medium may be a tangible medium that may contain or store a computer program for use by or in combination with an instruction execution system, device or equipment. A computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices or equipment, or any suitable combination of the foregoing. Alternatively, a computer-readable storage medium may be a machine-readable signal medium. A more specific example of a machine-readable storage medium may include an electrical connection based on one or more lines, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.
[0107] To provide interaction with a user, the systems and techniques described herein may be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and a pointing device (e.g., a mouse or trackball) through which the user can provide input to the electronic device. Other types of devices may also be used to provide interaction with the user; for example, the feedback provided to the user may be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user may be received in any form (including acoustic input, voice input, or tactile input).
[0108] The systems and techniques described herein may be implemented in a computing system that includes backend components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes frontend components (e.g., a user computer with a graphical user interface or a web browser through which a user can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such backend components, middleware components, or frontend components. The components of the system may be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: a local area network (LAN), a wide area network (WAN), a blockchain network, and the Internet.
[0109] A computing system may include a client and a server. The client and the server are generally remote from each other and usually interact through a communication network. The client and server relationship is generated by computer programs running on the corresponding computers and having a client-server relationship with each other. The server may be a cloud server, also known as a cloud computing server or cloud host, which is a host product in the cloud computing service system to solve the defects of difficult management and weak business scalability in traditional physical hosts and VPS services.
[0110] 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 the present invention can be executed in parallel, sequentially or in different orders, as long as the desired results of the technical solution of the present invention can be achieved, and this document does not limit this.
[0111] The above specific implementations do not constitute a limitation on the protection scope of the present invention. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions can be made according to design requirements and other factors. Any modification, equivalent substitution and improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims
1. A medical claims review method, It is characterized in that The method comprises: Obtaining pending medical claims information, and determining pending diseases, pending drugs, pending drug use indicator information, and current user profile information based on the pending medical claims information; Inputting the diseases to be reviewed, the drugs to be reviewed and the drug use index information to be reviewed into a preset basic medical model to obtain medical proper nouns to be reviewed; wherein the preset basic medical model is used to parse medical proper nouns; Determine historical user portrait information corresponding to the current user portrait information based on the disease to be reviewed, and determine the user portrait information to be reviewed based on the current user portrait information and the historical user portrait information; Determine the reference drug indications corresponding to the drug to be reviewed, input the disease to be reviewed, the drug to be reviewed, the medication indicator information to be reviewed, the medical term to be reviewed, the user portrait information to be reviewed, and the reference drug indications into a preset medical review model to obtain a medical review result.
2. The method according to claim 1, It is characterized in that The medical claims information to be reviewed includes the insurance policy information to be reviewed and the medical treatment information to be reviewed; Accordingly, the diseases to be reviewed, the drugs to be reviewed, the drug usage index information to be reviewed and the current user profile information are determined according to the medical claim information to be reviewed, including: Extracting the pending disease and the pending drug from the pending policy information; The medication index information to be reviewed and the current user portrait information are extracted from the medical information to be reviewed.
3. The method according to claim 2, It is characterized in that Before extracting the pending disease and the pending drug from the pending insurance policy information, the method further includes: Performing policy verification according to the policy information to be reviewed to obtain a verification result; wherein the verification result is pass or fail; Determine whether to initiate medical claims review based on the verification results.
4. The method according to claim 1, It is characterized in that The method further comprises: Determine historical user portrait information according to historical medical claims information, and store the historical user portrait information in a first database; wherein the historical user portrait information is used to describe user portrait information of historical candidate objects under historical candidate diseases; Accordingly, determining historical user portrait information corresponding to the current user portrait information based on the disease to be reviewed includes: Determine the historical candidate object in the first database that matches the current user portrait information as the target object; Determine the historical candidate disease in the first database that matches the disease to be reviewed as a target disease; The user portrait information under the target object and the target disease is obtained from the first database as the historical user portrait information corresponding to the current user portrait information.
5. The method according to claim 1, It is characterized in that The method further comprises: Determine the drug indication corresponding to each candidate drug according to the candidate insurance terms, and store the candidate drug and the drug indication corresponding to the candidate drug in a second database; Accordingly, the reference drug indications corresponding to the drug to be reviewed are determined, including: Determine the candidate drug in the second database that matches the drug to be reviewed as the target drug; The drug indication corresponding to the target drug is obtained from the second database as the reference drug indication corresponding to the drug to be reviewed.
6. The method according to any one of claims 1 to 5, It is characterized in that The method further comprises: Acquiring medical proper nouns as reference medical proper nouns, and storing the reference medical proper nouns in a third database; Accordingly, after obtaining the medical proper noun to be reviewed, the method further includes: Based on the reference medical terms stored in the third database, data normalization is performed on the diseases to be reviewed, the drugs to be reviewed, the drug use index information to be reviewed, the current user portrait information, and the medical terms to be reviewed.
7. The method according to claim 1, It is characterized in that The training process of the preset medical review model includes: Acquire historical manual review data, and perform data cleaning on the historical manual review data to obtain training data; Performing model training based on the training data to obtain a preset medical review model, and performing performance verification on the preset medical review model according to a model review result of the preset medical review model and a manual review result of the historical manual review data; If the preset medical review model fails the performance verification, the preset basic medical model is trained according to the training data, and the preset medical review model is tuned based on the training results of the preset basic medical model until the preset medical review model passes the performance verification.
8. A medical claims review device, It is characterized in that The device comprises: A pending medical claims information processing module is used to obtain pending medical claims information and determine pending diseases, pending drugs, pending drug use index information and current user portrait information based on the pending medical claims information; A module for determining medical terms to be reviewed, which is used to input the diseases to be reviewed, the drugs to be reviewed and the drug use index information to be reviewed into a preset basic medical model to obtain medical terms to be reviewed; wherein the preset basic medical model is used to parse medical terms; A module for determining user portrait information to be reviewed, used to determine historical user portrait information corresponding to the current user portrait information based on the disease to be reviewed, and to determine the user portrait information to be reviewed based on the current user portrait information and the historical user portrait information; The medical review result determination module is used to determine the reference drug indications corresponding to the drug to be reviewed, and input the disease to be reviewed, the drug to be reviewed, the medication indicator information to be reviewed, the medical term to be reviewed, the user portrait information to be reviewed, and the reference drug indications into a preset medical review model to obtain the medical review result.
9. An electronic device, It is characterized in that The electronic device comprises: at least one processor; and a memory communicatively connected to the at least one processor; wherein, The memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can perform the medical claim review method according to any one of claims 1 to 7.
10. A computer-readable storage medium, It is characterized in that The computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a processor to implement the medical claims review method according to any one of claims 1 to 7 when executed.
