Intelligent insurance claim settlement processing method and device based on AI, equipment and medium

Abstract

The invention discloses an intelligent insurance claim settlement processing method and device based on AI, equipment and a medium. The method comprises the following steps: constructing an insurance product digital model and a claim knowledge base; deploying a multi-terminal intelligent claim settlement channel, and receiving a claim settlement application and a multi-modal claim settlement material submitted by a user through the claim settlement channel; an intelligent material processing engine is adopted to carry out analysis processing on the multi-modal claim settlement material to obtain structured claim settlement data, and analysis processing comprises material classification, image optimization, key information identification and medical text semantic analysis; based on the insurance product digital model, the claim settlement knowledge base and the structured claim settlement data, claim settlement auditing is executed through an AI intelligent auditing decision flow, and a claim settlement auditing result is generated; and the claim settlement audit result and the claim settlement progress information are fed back to the user through the multi-terminal intelligent claim settlement channel, and the claim settlement operation is executed, so that the problems of difficulty in material identification, dependence on manpower in the process, low efficiency and the like in traditional insurance claim settlement can be solved, and intelligent processing of the whole process of insurance claim settlement is realized.

Description

Technical Field

[0001] This invention relates to the fields of artificial intelligence technology and insurance claims technology, and more particularly to an AI-based intelligent insurance claims processing method. Background Technology

[0002] In traditional insurance claims processes, insured individuals need to submit more than ten types of documents. The process is complex and highly reliant on manual labor, resulting in a heavy burden of documentation, long claims processing times, high labor costs, and a lack of transparency. Existing solutions cannot accurately identify non-standard medical documents, comprehensively verify medical indications, or automate the claims process. Users need to repeatedly supplement materials, insurance companies rely on manual review for over 70% of cases, individual case processing takes over 30 minutes, and the overall claims processing cycle takes 3-5 business days. This negatively impacts user experience and increases operating costs for insurance companies. Therefore, there is an urgent need for an AI-based intelligent insurance claims processing method to address these issues. Summary of the Invention

[0003] This invention provides an AI-based intelligent insurance claims processing method, device, equipment, and medium to solve problems such as difficulty in material identification, reliance on manual processes, and low efficiency in traditional insurance claims, thereby achieving intelligent processing of the entire insurance claims process.

[0004] According to one aspect of the present invention, an AI-based intelligent insurance claims processing method is provided, comprising:

[0005] Build digital models of insurance products and claims knowledge bases;

[0006] Deploy a multi-terminal intelligent claims channel to receive claims applications and multimodal claims materials submitted by users through the multi-terminal intelligent claims channel;

[0007] The multimodal claims materials are parsed and processed using an intelligent materials processing engine to obtain structured claims data. The parsing and processing includes at least one of the following: material classification, image optimization, key information identification, and medical text semantic analysis.

[0008] Based on the digital model of the insurance product, the claims knowledge base, and the structured claims data, the claims review is executed through an AI-powered intelligent review decision flow to generate claims review results.

[0009] The multi-terminal intelligent claims channel provides users with feedback on the claims review results and claims progress information, and executes the claims operation based on the claims review results.

[0010] According to another aspect of the present invention, an AI-based intelligent insurance claims processing device is provided, comprising:

[0011] The knowledge base construction module is used to build digital models of insurance products and claims knowledge bases;

[0012] The claims channel deployment module is used to deploy a multi-terminal intelligent claims channel, through which claims applications and multimodal claims materials submitted by users are received;

[0013] The material analysis and processing module is used to analyze and process the multimodal claims materials using an intelligent material processing engine to obtain structured claims data. The analysis and processing includes at least one of material classification, image optimization, key information identification, and medical text semantic analysis.

[0014] The review result determination module is used to perform a claims review based on the digital model of the insurance product, the claims knowledge base, and the structured claims data, and to generate a claims review result by executing the claims review through an AI intelligent review decision flow.

[0015] The claims result execution module is used to provide users with feedback on the claims review results and claims progress information through the multi-terminal intelligent claims channel, and to execute claims operations based on the claims review results.

[0016] According to another aspect of the present invention, an electronic device is provided, the electronic device comprising:

[0017] At least one processor;

[0018] and memory that is communicatively connected to at least one processor;

[0019] The memory stores a computer program that can be executed by at least one processor, which is then executed by the at least one processor to enable the at least one processor to execute the AI-based intelligent insurance claims processing method according to any embodiment of the present invention.

[0020] According to another aspect of the present invention, a computer-readable storage medium is provided, which stores computer instructions for causing a processor to execute and implement an AI-based intelligent insurance claims processing method according to any embodiment of the present invention.

[0021] The technical solution of this invention addresses the problems of difficult material identification, reliance on manual processes, and low efficiency in traditional insurance claims by constructing a digital model of insurance products and a claims knowledge base; deploying multi-terminal intelligent claims channels to receive user-submitted claims applications and multimodal claims materials; employing an intelligent material processing engine to parse and process the multimodal claims materials to obtain structured claims data, including material classification, image optimization, key information identification, and medical text semantic analysis; based on the digital model of insurance products, the claims knowledge base, and the structured claims data, executing claims review through an AI intelligent review decision flow to generate claims review results; and providing users with feedback on claims review results and claims progress information through the multi-terminal intelligent claims channels, and executing claims operations. This solution achieves intelligent processing of the entire insurance claims process.

[0022] It should be understood that the description in this section is not intended to identify key or essential features of the embodiments of the present invention, nor is it intended to limit the scope of the invention. Other features of the invention will become readily apparent from the following description. Attached Figure Description

[0023] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0024] Figure 1 A flowchart illustrating an AI-based intelligent insurance claims processing method provided in an embodiment of the present invention;

[0025] Figure 2 A flowchart illustrating another AI-based intelligent insurance claims processing method provided in an embodiment of the present invention;

[0026] Figure 3 A schematic diagram of the structure of an AI-based intelligent insurance claims processing device provided in an embodiment of the present invention;

[0027] Figure 4 A schematic diagram of the structure of an electronic device for implementing the AI-based intelligent insurance claims processing method of this invention. Detailed Implementation

[0028] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0029] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0030] Figure 1 This is a flowchart illustrating an AI-based intelligent insurance claims processing method provided in an embodiment of the present invention. This embodiment is applicable to insurance claims review. The method can be executed by an AI-based intelligent insurance claims processing device, which can be implemented in hardware and / or software and can be configured in an electronic device. Figure 1 As shown, the method specifically includes the following steps:

[0031] S110. Construct digital models of insurance products and a claims knowledge base.

[0032] Among them, the digital model of insurance products refers to the model that transforms the scope of liability, compensation conditions and other contents of insurance products into digital form, including product liability matrix and structured liability tree; the claims knowledge base can be understood as a database that stores information such as claims-related rules and logic, including diagnosis-treatment logic, clinical rationality rules, etc.

[0033] S120. Deploy a multi-terminal intelligent claims channel to receive claims applications and multimodal claims materials submitted by users through the multi-terminal intelligent claims channel.

[0034] Among them, the multi-terminal intelligent claims channel can be understood as the claims entry point deployed on different terminals, including WeChat mini-program, H5 and APP; multimodal claims materials can refer to the various types of claims materials submitted by users, including but not limited to medical bills, medical records, identity documents, etc.

[0035] S130. The multimodal claims materials are analyzed and processed using an intelligent material processing engine to obtain structured claims data.

[0036] Among them, the intelligent material processing engine can be a processing module with the ability to classify materials, optimize images, identify key information and analyze medical text semantics; parsing processing includes at least one of material classification, image optimization, key information identification and medical text semantic analysis; structured claims data can refer to claims data that includes diagnostic information, treatment plans and other content, and is presented in a standardized format after parsing processing.

[0037] S140. Based on the digital model of the insurance product, the claims knowledge base, and the structured claims data, the claims review is executed through AI intelligent review decision flow to generate claims review results.

[0038] Among them, the AI ​​intelligent review decision flow can be a process that executes claims review based on AI technology, and can complete operations such as material verification and risk identification; the claims review result can refer to the result of approval or disapproval generated after the review, including the amount of compensation or the reason for disapproval; the claims progress information refers to the progress of each node in the claims process, and claims operations include triggering compensation instructions, re-reviewing materials, etc.

[0039] S150. Feedback the claim review results and claim progress information to the user through the multi-terminal intelligent claim channel, and execute the claim operation according to the claim review results.

[0040] In this embodiment of the invention, a digital model of insurance products and a claims knowledge base can be built first to provide standards and basis for subsequent claims review; then, a multi-terminal intelligent claims channel can be deployed to receive claims applications and multimodal claims materials submitted by users, so as to meet the claims needs of users on different terminals.

[0041] Furthermore, an intelligent material processing engine is employed to perform at least one of the following operations on multimodal claim materials: material classification, image optimization, key information identification, and medical text semantic analysis, transforming unstructured materials into structured claim data. Subsequently, based on the constructed digital model of the insurance product, the claim knowledge base, and the obtained structured claim data, an AI-powered intelligent review decision flow is used to execute the claim review and generate the claim review results. Finally, the claim review results and progress information are fed back to the user through a multi-terminal intelligent claim channel, and corresponding claim operations are executed according to the review results, such as triggering a payment instruction for users who have passed the review, and re-analyzing and reviewing cases with supplementary materials submitted by users.

[0042] The advantage of this is that it enables the entire claims process, from material collection to result feedback, to be automated, reducing human intervention, shortening the claims cycle, improving claims efficiency and user experience, and reducing the labor costs of insurance companies.

[0043] In some optional embodiments, the digital model of the insurance product includes a product liability matrix and a structured liability tree, and the claims knowledge base includes diagnosis-treatment logic, clinical rationality rules, material integrity rules, and risk interception rules.

[0044] The construction of the digital model and claims knowledge base for insurance products includes: inputting insurance product information, product-related agreements, and corresponding product rules to build a basic framework for the claims knowledge base; calling an actuarial factor engine to calculate product liability-related parameters and generate a product liability matrix; using an insurance clause parser to decompose insurance clauses, extract liability elements, and construct a structured liability tree; integrating the structured liability tree with the product liability matrix to obtain the digital model of the insurance product; collecting the diagnosis-treatment logic provided by professional physicians, the clinical rationality rules defined by claims review specialists, the material completeness rules, and the risk interception rules; and inputting the diagnosis-treatment logic, clinical rationality rules, material completeness rules, and risk interception rules into the basic framework of the claims knowledge base to obtain the overall construction of the claims knowledge base.

[0045] Among them, the product liability matrix can be generated by calculating product liability-related parameters through an actuarial factor engine, and is used to clarify the scope of insurance product liability; the structured liability tree can be constructed by disassembling insurance clauses and extracting liability elements using an insurance clause parser, and is used to present the hierarchical relationship of insurance liability; the diagnosis-treatment logic can be the logic provided by professional physicians regarding disease diagnosis standards and treatment plan selection; the clinical rationality rules can be the rules defined by claims review specialists to determine whether treatment plans, medications, etc., comply with clinical norms.

[0046] Material completeness rules can be understood as rules used to determine whether claim materials cover all required elements; risk interception rules can be understood as rules used to identify risks such as fraud in claim cases; insurance product information includes basic content such as product name and coverage period; product-related agreements are contract texts related to insurance products; product rules are rules such as compensation standards and exclusions for insurance products; the basic framework of the claims knowledge base can be the basic structure of the knowledge base built after inputting insurance product information, agreements, and rules; the actuarial factor engine can be a tool used to calculate liability-related parameters of insurance products; and the insurance clause parser can be a tool used to deconstruct insurance clauses and extract liability elements.

[0047] Specifically, insurance product information, related agreements, and corresponding product rules can be entered to build a basic framework for the claims knowledge base; the actuarial factor engine can be invoked to calculate product liability-related parameters in conjunction with the scope of liability of the insurance product, generating a product liability matrix; the insurance clause parser can be used to decompose the insurance clauses, extract liability elements such as liability type, compensation conditions, and exemption circumstances, construct a structured liability tree based on these elements, and integrate the structured liability tree with the generated product liability matrix to obtain a digital model of the insurance product.

[0048] Furthermore, the diagnostic-treatment logic provided by professional physicians, as well as the clinical rationale rules, document completeness rules, and risk interception rules defined by claims review specialists, are collected. The collected diagnostic-treatment logic and various rules are then entered into the established claims knowledge base framework, completing the overall construction of the claims knowledge base. This ensures that the digital model of insurance products accurately reflects insurance liability, and that the claims knowledge base provides comprehensive review evidence, thereby improving the accuracy of the review process.

[0049] In some optional embodiments, the deployment of the multi-terminal intelligent claims channel includes: developing and deploying claims entry points on a WeChat mini-program, an H5 page, and an APP, forming the multi-terminal intelligent claims channel; configuring material uploading, application filling, and progress query function modules for the WeChat mini-program and H5 pages, and integrating biometric verification function for the APP page.

[0050] Among them, the WeChat mini-program claims entry can be a claims function entry deployed in a WeChat mini-program; the H5 claims entry can be a claims function entry developed based on H5 technology that can be accessed in a browser; the APP claims entry can be a claims function entry deployed in a dedicated APP; the material upload function module is a function module for users to upload multimodal claims materials; the application filling function module is a function module for users to fill in claims application information; the progress query function module is a function module for users to query the progress of the claims process; and the biometric verification function is a function for verifying the user's identity through the user's biometrics (such as fingerprints, faces).

[0051] Specifically, claims entry points can be developed on WeChat mini-programs, H5 pages, and mobile apps, integrating these three platforms to form a multi-terminal intelligent claims channel covering commonly used user devices. Then, modules for uploading materials, filling out applications, and checking claims progress can be configured for the WeChat mini-program and H5 pages, allowing users to easily submit claims materials, fill out application information, and check claims progress on both devices. Simultaneously, biometric verification functionality can be integrated into the mobile app to verify user identity and enhance the security of claims processing on the app. This provides users with diverse claims terminal options to meet different user habits, while biometric verification ensures user identity security and increases user trust in the claims process.

[0052] The technical solution of this invention addresses the problems of difficult material identification, reliance on manual processes, and low efficiency in traditional insurance claims by constructing a digital model of insurance products and a claims knowledge base; deploying multi-terminal intelligent claims channels to receive user-submitted claims applications and multimodal claims materials; using an intelligent material processing engine to parse and process the multimodal claims materials to obtain structured claims data, including material classification, image optimization, key information identification, and medical text semantic analysis; based on the digital model of insurance products, the claims knowledge base, and the structured claims data, executing claims review through an AI intelligent review decision flow to generate claims review results; and providing users with feedback on claims review results and claims progress information through the multi-terminal intelligent claims channels, and executing claims operations. This solution achieves intelligent processing of the entire insurance claims process.

[0053] Figure 2 This is a flowchart illustrating another AI-based intelligent insurance claims processing method provided by an embodiment of the present invention. Based on the above embodiments, this embodiment further optimizes the process of parsing and processing multimodal claims materials and the claims review process. Technical terms that are the same as or corresponding to those in the above embodiments will not be repeated here. Figure 2 As shown, the method specifically includes the following steps:

[0054] S210. Construct digital models of insurance products and a claims knowledge base.

[0055] S220. Deploy a multi-terminal intelligent claims channel to receive claims applications and multimodal claims materials submitted by users through the multi-terminal intelligent claims channel.

[0056] S230. The multimodal claims materials are analyzed and processed using an intelligent material processing engine to obtain structured claims data.

[0057] The parsing process includes at least one of material classification, image optimization, key information identification, and medical text semantic analysis.

[0058] In some embodiments, the step of using an intelligent material processing engine to parse and process the multimodal claim materials to obtain structured claim data includes: classifying the multimodal claim materials using large-model multimodal feature recognition technology, distinguishing between medical invoices, medical records, identity documents, and financial account verification types, and performing automatic correction on tilted or blurry image materials; extracting key information from the medical invoices using OCR recognition technology, processing the medical records using NLP semantic analysis technology, extracting diagnostic information, treatment plans, and disease progression information, and achieving mapping between diagnostic information and standard codes, as well as normalization of drug names; cross-validating the results of the OCR recognition and the results of the NLP semantic analysis to correct recognition errors, and generating the structured claim data, wherein the structured claim data includes at least diagnostic information, treatment plans, disease progression information, identity verification information, and financial account verification information.

[0059] Among them, large-scale multimodal feature recognition technology can be understood as a technology capable of recognizing features of various types of materials, classifying materials, and optimizing images; medical bills can be receipts for expenses obtained by users after seeking medical treatment, such as invoices and expense details; medical records can be understood as documents that record a user's condition, diagnosis, and treatment process, such as medical records and discharge summaries; identity verification is material used to prove a user's identity, such as an ID card; financial account verification is account verification material used to receive claims payments, such as bank cards, etc.

[0060] OCR (Optical Character Recognition) technology is used to extract key information from invoices, including but not limited to information such as drug name, specifications, dosage, unit price, and invoice date in medical invoices; NLP (Natural Language Processing) semantic analysis technology can be used to process medical records, extract relevant information, and implement encoding mapping and name normalization; standard encoding is used to unify the encoding of diagnostic information, such as ICD-10 encoding; drug name normalization can be the process of converting the brand name, English name, etc. of drugs into a common name;

[0061] Cross-validation can be an operation that verifies the results of OCR recognition and NLP semantic analysis; recognition error refers to the information extraction error that occurs during the OCR recognition or NLP semantic analysis process; identity verification information is user identity information extracted from identity verification materials; financial account verification information is account information extracted from financial account verification materials.

[0062] Specifically, large-scale multimodal feature recognition technology can be used to classify received multimodal claims materials, distinguishing between types such as medical invoices, medical records, identity documents, and financial account verification. Simultaneously, image-based materials with tilting or blurring issues after classification are automatically corrected to ensure clarity and recognizability. Then, OCR technology is used to extract key information from medical invoices, and NLP semantic analysis technology is used to process medical records, extracting diagnostic information, treatment plans, and disease progression information. The diagnostic information is mapped to standard codes, and drug names are normalized. Finally, the key information results obtained from OCR recognition are cross-validated with the diagnostic and treatment information results obtained from NLP semantic analysis to correct any recognition errors, generating structured claims data containing diagnostic information, treatment plans, disease progression information, identity document information, and financial account verification information. The advantages of this approach are that it achieves accurate classification and optimization of multimodal claims materials, ensures accurate extraction of key information, generates standardized structured claims data, provides high-quality data support for AI-powered intelligent review, and improves review accuracy.

[0063] In this embodiment of the invention, the claims review includes verification of material integrity, verification of indication matching, verification of consistency of disease progression logic, and verification of risk interception.

[0064] S240. Based on the material integrity rules, verify whether the structured claims data covers all the elements required for claims, and generate a material completion prompt if any are missing.

[0065] S250. Based on the product liability matrix and the structured liability tree, match the diagnostic information and treatment plan in the structured claims data with the scope of insurance liability to verify the compliance of the indications.

[0066] S260. Analyze the disease progression information in the structured claims data using a preset disease evolution model to verify the consistency of the disease progression logic.

[0067] Among them, material integrity verification refers to judging whether the structured claims data contains all the elements required for claims based on material integrity rules; indication matching verification can be based on the product liability matrix and structured liability tree to judge whether the diagnostic information and treatment plan are within the scope of insurance liability; disease progression logic consistency verification can be based on a preset disease evolution model to judge whether the treatment logic in the disease progression information is reasonable; and risk interception verification can be based on risk interception rules to identify whether there are risks in the structured claims data.

[0068] The scope of insurance liability can be the scope of compensation liability that the insurance company must bear, as clearly defined in the digital model of the insurance product; the preset disease evolution model can be a pre-trained model used to analyze disease progression and determine whether the treatment logic is reasonable; risk identification is the operation of determining whether there are risks such as fraud in a claim case; high risk refers to situations with a high probability of fraud or other situations that do not meet the requirements for claims.

[0069] Specifically, based on the material completeness rules in the claims knowledge base, it can be verified whether the structured claims data covers all the elements required for claims. If there are missing elements, a material completion prompt will be generated to notify the user to supplement the materials. Then, based on the product liability matrix and structured liability tree in the digital model of insurance products, the diagnostic information and treatment plan in the structured claims data will be matched with the scope of insurance liability to verify the compliance of indications and determine whether the diagnosis and treatment are within the scope of insurance liability.

[0070] Furthermore, a pre-defined disease evolution model can be used to analyze the disease progression information in structured claims data, verifying the logical consistency of the disease progression and determining whether the treatment plan matches the disease progression. Then, risk identification can be performed on the structured claims data based on risk interception rules in the claims knowledge base. If a high-risk situation is identified, the case is marked as a case requiring manual review. In addition, for cases that pass all the above verifications, the compensation amount can be calculated according to the scope of insurance liability and relevant rules, generating an approved result; for cases that fail any verification, an unapproved result including the reason for failure is generated. The approved and unapproved results together constitute the claims review result. Through multi-dimensional verification operations, claims cases are comprehensively reviewed to ensure the accuracy of the review results. At the same time, risk identification reduces unreasonable payouts and lowers the risk for insurance companies, while manual review handles complex situations, balancing automation and review accuracy.

[0071] In some optional embodiments, the step of analyzing the disease progression information in the structured claims data using a preset disease evolution model and verifying the consistency of treatment logic includes: extracting pathological indicators, diagnostic results, treatment plans, and medication records from the structured claims data to construct a medical disease progression chain with a time axis; inputting the medical disease progression chain into the preset disease evolution model to obtain a disease progression prediction trajectory; comparing the matching degree between the actual treatment information in the medical disease progression chain and the disease progression prediction trajectory, and determining that the treatment logic is inconsistent when the matching degree is lower than a threshold.

[0072] The medical disease progression chain can be a chain that presents the user's disease progression and treatment process in chronological order, based on pathological indicators, diagnostic results, treatment plans, and medication records. The disease progression prediction trajectory can be the disease development trajectory predicted by a preset disease evolution model based on the medical disease progression chain. The actual treatment information can be the treatment plan, medication, and other information actually received by the user recorded in the medical disease progression chain. Inconsistent treatment logic can be understood as a situation where the actual treatment information does not match the disease progression prediction trajectory, and the treatment plan does not conform to the disease development pattern.

[0073] Specifically, information related to the user's condition and treatment, such as pathological indicators, diagnostic results, treatment plans, and medication records, is extracted from structured claims data. A timeline-based medical progression chain is constructed based on the chronological order of this information to present the user's disease progression and treatment process. This constructed medical progression chain is then input into a pre-defined disease evolution model. The model analyzes the information in the medical progression chain and outputs a predicted disease progression trajectory. Furthermore, the matching degree between the actual treatment information in the medical progression chain and the predicted disease progression trajectory output by the model is compared. If the matching degree is lower than a preset threshold, the treatment logic is deemed inconsistent, and the case requires further review. By analyzing the treatment logic through the pre-defined disease evolution model, it ensures that the treatment plan conforms to the disease development pattern, reduces erroneous payouts due to unreasonable treatment logic, and improves the professionalism and accuracy of claims review.

[0074] S270. Based on the risk interception rules, perform risk identification on the structured claims data, and mark high-risk cases as cases requiring manual review.

[0075] S280. Calculate the compensation amount for cases that pass the verification and generate the verification approval result; generate the verification failure result including the reason for failure for cases that fail the verification.

[0076] The claims review results include approved and unapproved results.

[0077] S290. Feedback the claim review results and claim progress information to the user through the multi-terminal intelligent claim channel, and execute the claim operation according to the claim review results.

[0078] In some optional embodiments, the step of feeding back the claim review results and claim progress information to the user through the multi-terminal intelligent claims channel, and executing the claim operation based on the claim review results, includes: pushing the claim review results through the multi-terminal intelligent claims channel; synchronizing payment-related information when the review is passed; explaining the reasons and supplementary requirements when the review is failed; updating the claim progress information in real time; pushing process node change notifications through the multi-terminal intelligent claims channel; triggering payment instructions and sending them to the account notification for cases that have passed the review; and for cases where the user provides supplementary materials, re-executing the intelligent parsing and review process as new multimodal claim materials.

[0079] Specifically, the system can push claim review results to users through a multi-terminal intelligent claims channel. If the review is approved, the user will be informed of the compensation amount, expected arrival time, and other relevant compensation information. If the review is not approved, the user will be informed of the reasons for the failure and the supplementary materials or information required. The system will then update the progress information of the claims process in real time. Whenever the claims process moves from one stage to the next, such as from the material review stage to the compensation stage, the system will push a process stage change notification to the user through the multi-terminal intelligent claims channel, so that the user can keep abreast of the claims progress.

[0080] For approved cases, a payment instruction can be triggered to the financial system. Once the claim payment arrives in the user's account, a notification will be sent to the user through a multi-terminal intelligent claims channel. For cases that fail the review and the user submits supplementary materials as required, these supplementary materials will be treated as new multimodal claims materials. The intelligent material processing engine will then perform intelligent analysis to obtain new structured claims data before re-executing the AI ​​intelligent review process. This allows for timely feedback of claims-related information to users, improving the transparency of the claims process, reducing user questions arising from unclear information, and ensuring that users have the opportunity to complete their materials and receive reasonable compensation by re-reviewing supplementary materials, thus enhancing the user experience.

[0081] The technical solution of this invention addresses the problems of difficult material identification, reliance on manual processes, and low efficiency in traditional insurance claims by constructing a digital model of insurance products and a claims knowledge base; deploying multi-terminal intelligent claims channels to receive user-submitted claims applications and multimodal claims materials; using an intelligent material processing engine to parse and process the multimodal claims materials to obtain structured claims data, including material classification, image optimization, key information identification, and medical text semantic analysis; based on the digital model of insurance products, the claims knowledge base, and the structured claims data, executing claims review through an AI intelligent review decision flow to generate claims review results; and providing users with feedback on claims review results and claims progress information through the multi-terminal intelligent claims channels, and executing claims operations. This solution achieves intelligent processing of the entire insurance claims process.

[0082] Figure 3 This is a schematic diagram of the structure of an AI-based intelligent insurance claims processing device provided in an embodiment of the present invention. Figure 3 As shown, the device includes:

[0083] Knowledge base construction module 310 is used to build digital models of insurance products and claims knowledge bases;

[0084] The claims channel deployment module 320 is used to deploy a multi-terminal intelligent claims channel, through which the user submits claims applications and multimodal claims materials;

[0085] The material analysis and processing module 330 is used to analyze and process the multimodal claims materials using an intelligent material processing engine to obtain structured claims data. The analysis and processing includes at least one of material classification, image optimization, key information recognition, and medical text semantic analysis.

[0086] The review result determination module 340 is used to perform a claims review based on the digital model of the insurance product, the claims knowledge base, and the structured claims data, and to generate a claims review result by executing the claims review through an AI intelligent review decision flow.

[0087] The claims result execution module 350 is used to provide feedback on the claims review results and claims progress information to the user through the multi-terminal intelligent claims channel, and to execute claims operations based on the claims review results.

[0088] In some embodiments, the digital model of the insurance product includes a product liability matrix and a structured liability tree, and the claims knowledge base includes diagnosis-treatment logic, clinical rationality rules, material integrity rules, and risk interception rules.

[0089] The knowledge base construction module 310 is specifically used for:

[0090] Enter insurance product information, product-related agreements, and corresponding product rules to build the basic framework of the claims knowledge base;

[0091] The actuarial factor engine is invoked to calculate product liability-related parameters and generate a product liability matrix.

[0092] An insurance clause parser is used to disassemble the insurance clauses, extract liability elements, and construct a structured liability tree. The structured liability tree is then integrated with the product liability matrix to obtain a digital model of the insurance product.

[0093] Collect the diagnostic-treatment logic provided by professional physicians, the clinical rationality rules defined by claims review specialists, the material completeness rules, and the risk interception rules;

[0094] The diagnosis-treatment logic, clinical rationality rules, material integrity rules, and risk interception rules are entered into the basic framework of the claims knowledge base to obtain the overall construction of the claims knowledge base.

[0095] In some embodiments, the claims channel deployment module 320 is specifically used for:

[0096] Develop and deploy claims entry points on WeChat mini-program, H5, and APP to form the multi-terminal intelligent claims channel;

[0097] Configure modules for uploading materials, filling out applications, and querying progress for the WeChat mini-program and H5 platforms, and integrate biometric verification functionality for the APP platform.

[0098] In some embodiments, the material analysis processing module 330 is specifically used for:

[0099] The multimodal feature recognition technology of large model is used to classify the multimodal claims materials, distinguish between medical bills, medical records, identity documents and financial account certificates, and perform automatic correction on tilted and blurry image materials;

[0100] Key information in the medical invoices is extracted using OCR recognition technology, and medical records are processed using NLP semantic analysis technology to extract diagnostic information, treatment plans, and disease progression information, thereby achieving the mapping of diagnostic information to standard codes and the normalization of drug names.

[0101] The results of the OCR recognition and the NLP semantic analysis are cross-validated to correct recognition errors and generate the structured claims data. The structured claims data includes at least diagnostic information, treatment plan, disease course information, identity verification information, and financial account verification information.

[0102] In some embodiments, the claims review includes verification of the completeness of materials, verification of indication matching, verification of consistency of medical course logic, and verification of risk interception.

[0103] The audit result determination module 340 includes:

[0104] The material integrity verification submodule is used to verify whether the structured claims data covers all the elements required for claims based on the material integrity rules, and to generate a material completion prompt when there are any missing elements.

[0105] The indication matching and verification submodule is used to match the diagnostic information and treatment plan in the structured claims data with the scope of insurance liability based on the product liability matrix and the structured liability tree, and to verify the conformity of the indications.

[0106] The disease progression logic consistency verification submodule is used to analyze the disease progression information in the structured claims data using a preset disease evolution model and verify the consistency of the disease progression logic.

[0107] The risk interception and verification submodule is used to identify risks in the structured claims data based on the risk interception rules, and mark high-risk cases as cases requiring manual review.

[0108] The review result determination submodule is used to calculate the compensation amount for cases that pass the verification and generate a review pass result, and to generate a review fail result including the reason for the failure for cases that fail the verification. The claims review result includes a review pass result and a review fail result.

[0109] In some embodiments, the disease course logic consistency verification submodule is specifically used for:

[0110] Pathological indicators, diagnostic results, treatment plans, and medication records are extracted from the structured claims data to construct a medical disease process chain with a timeline.

[0111] The medical disease progression chain is input into the preset disease evolution model to obtain the disease progression prediction trajectory;

[0112] The matching degree between the actual treatment information in the medical disease process chain and the predicted trajectory of disease progression is compared. If the matching degree is lower than a threshold, the treatment logic is determined to be inconsistent.

[0113] In some embodiments, the claims result execution module 350 is specifically used for:

[0114] The claims review results are pushed through the multi-terminal intelligent claims channel. When the review is approved, the relevant compensation information is synchronized. When the review is not approved, the reasons and supplementary requirements are explained.

[0115] Real-time updates of claims progress information, and push notifications of process node changes through the multi-terminal intelligent claims channel;

[0116] For cases that have passed the review, a claim instruction is triggered and sent to the account notification. For cases where the user provides supplementary materials, the supplementary materials are treated as new multimodal claim materials and the intelligent analysis and review process is re-executed.

[0117] The AI-based intelligent insurance claims processing device provided in this embodiment of the invention can execute the AI-based intelligent insurance claims processing method provided in any embodiment of the invention, and has the corresponding functional modules and beneficial effects of the method.

[0118] Figure 4This is a schematic diagram of the structure of an electronic device for implementing the AI-based intelligent insurance claims processing method according to embodiments of the present invention. The electronic device is intended to represent various forms of digital computers, such as laptops, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframes, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital assistants, cellular phones, smartphones, wearable devices (e.g., helmets, glasses, watches, etc.), and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the invention described and / or claimed herein.

[0119] like Figure 4 As shown, the electronic device 10 includes at least one processor 11 and a memory, such as a read-only memory (ROM) 12 or a random access memory (RAM) 13, communicatively connected to the at least one processor 11. The memory stores computer programs executable by the at least one processor. The processor 11 can perform various appropriate actions and processes based on the computer program stored in the ROM 12 or loaded from storage unit 18 into the RAM 13. The RAM 13 can also store various programs and data required for the operation of the electronic device 10. The processor 11, ROM 12, and RAM 13 are interconnected via a bus 14. An input / output (I / O) interface 15 is also connected to the bus 14.

[0120] Multiple components in electronic device 10 are connected to I / O interface 15, including: input unit 16, such as keyboard, mouse, etc.; output unit 17, such as various types of displays, speakers, etc.; storage unit 18, such as disk, optical disk, etc.; and communication unit 19, such as network card, modem, wireless transceiver, etc. Communication unit 19 allows electronic device 10 to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.

[0121] Processor 11 can be a variety of general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of processor 11 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special-purpose artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. Processor 11 performs the various methods and processes described above, such as AI-based intelligent insurance claims processing methods.

[0122] In some embodiments, the AI-based intelligent insurance claims processing method can be implemented as a computer program tangibly contained in a computer-readable storage medium, such as storage unit 18. In some embodiments, part or all of the computer program can be loaded and / or installed on electronic device 10 via ROM 12 and / or communication unit 19. When the computer program is loaded into RAM 13 and executed by processor 11, one or more steps of the AI-based intelligent insurance claims processing method described above can be performed. Alternatively, in other embodiments, processor 11 can be configured to perform the AI-based intelligent insurance claims processing method by any other suitable means (e.g., by means of firmware).

[0123] 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-a-chip (SoCs), complex programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various implementations may include: implementations in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which may be a dedicated or general-purpose programmable processor, capable of receiving data and instructions from a storage system, at least one input device, and at least one output device, and transmitting data and instructions to the storage system, the at least one input device, and the at least one output device.

[0124] Computer programs used to implement 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, such that when executed by the processor, the computer programs cause the functions / operations specified in the flowcharts and / or block diagrams to be performed. The computer programs may be executed entirely on a machine, partially on a machine, or as a standalone software package, partially on a machine and partially on a remote machine, or entirely on a remote machine or server.

[0125] In the context of this invention, a computer-readable storage medium can be a tangible medium that may contain or store a computer program for use by or in conjunction with an instruction execution system, apparatus, or device. A computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination thereof. Alternatively, a computer-readable storage medium may be a machine-readable signal medium. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fibers, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof.

[0126] To provide interaction with a user, the systems and techniques described herein can 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 pointing device (e.g., a mouse or trackball) through which the user provides input to the electronic device. Other types of devices can also be used to provide interaction with the user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including sound input, voice input, or tactile input).

[0127] The systems and technologies described herein can be implemented in computing systems that include backend components (e.g., as data servers), or middleware components (e.g., application servers), or frontend components (e.g., user computers with graphical user interfaces or web browsers through which users can interact with implementations of the systems and technologies described herein), or any combination of such backend, middleware, or frontend components. The components of the system can be interconnected via digital data communication of any form or medium (e.g., communication networks). Examples of communication networks include local area networks (LANs), wide area networks (WANs), blockchain networks, and the Internet.

[0128] A computing system can include clients and servers. Clients and servers are generally located far apart and typically interact through communication networks. The client-server relationship is created by computer programs running on the respective computers and having a client-server relationship with each other. The server can be a cloud server, also known as a cloud computing server or cloud host, which is a hosting product within the cloud computing service system to address the shortcomings of traditional physical hosts and VPS services, such as high management difficulty and weak business scalability.

[0129] It should be understood that the various forms of processes shown above can be used, with steps reordered, added, or deleted. 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 of this invention can be achieved, and this is not limited herein.

[0130] The specific embodiments described above do not constitute a limitation on the scope of protection of this invention. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this invention should be included within the scope of protection of this invention.

Claims

1. An AI-based intelligent insurance claim processing method, characterized in that, The method comprises the following steps: constructing an insurance product digital model and a claim knowledge base; deploying a multi-terminal intelligent claim channel, receiving a claim application and multi-modal claim materials submitted by a user through the multi-terminal intelligent claim channel; using an intelligent material processing engine to analyze and process the multi-modal claim materials to obtain structured claim data, wherein the analysis and processing includes at least one of material classification, image optimization, key information identification, and medical text semantic analysis; based on the insurance product digital model, the claim knowledge base, and the structured claim data, performing claim auditing through an AI intelligent auditing decision flow to generate a claim auditing result; feeding back the claim auditing result and claim progress information to the user through the multi-terminal intelligent claim channel, and performing claim operations according to the claim auditing result.

2. The method of claim 1, wherein, The insurance product digital model includes a product liability matrix and a structured liability tree, and the claim knowledge base includes diagnosis-treatment logic, clinical reasonableness rules, material integrity rules, and risk interception rules; The construction of the insurance product digital model and the claim knowledge base comprises: entering insurance product information, product-related agreements, and corresponding product rules to build a basic framework of the claim knowledge base; calling an actuarial factor engine to calculate product liability correlation parameters to generate a product liability matrix; using an insurance clause parser to disassemble insurance clauses, extract liability elements, and construct a structured liability tree, integrating the structured liability tree and the product liability matrix to obtain the insurance product digital model; collecting the diagnosis-treatment logic provided by professional physicians, the clinical reasonableness rules defined by claim auditing officers, the material integrity rules, and the risk interception rules; entering the diagnosis-treatment logic, clinical reasonableness rules, material integrity rules, and risk interception rules into the basic framework of the claim knowledge base to obtain the overall construction of the claim knowledge base.

3. The method of claim 1, wherein, The deployment of the multi-terminal intelligent claim channel comprises: developing and deploying WeChat mini-program, H5, and APP claim entry to form the multi-terminal intelligent claim channel; configuring material uploading, application filling, and progress inquiry function modules for the WeChat mini-program and H5, and integrating biometric verification functions for the APP.

4. The method of claim 2, wherein, The use of an intelligent material processing engine to analyze and process the multi-modal claim materials to obtain structured claim data comprises: using large model multi-modal feature recognition technology to classify the multi-modal claim materials, distinguishing between medical bills, medical records, identification, and financial account proof types, and performing automatic correction on tilted and blurred image materials; using OCR recognition technology to extract key information from the medical bills, using NLP semantic analysis technology to process the medical records to extract diagnosis information, treatment plans, and medical history information, realizing the mapping of diagnosis information and standard codes, and drug name normalization processing; cross-checking the results of the OCR recognition and the NLP semantic analysis to correct recognition errors to generate the structured claim data, wherein the structured claim data at least includes diagnosis information, treatment plans, medical history information, identification information, and financial account proof information.

5. The method of claim 4, wherein, The claim settlement audit includes material integrity verification, indication matching verification, disease course logical consistency verification, and risk interception verification. The AI intelligent audit decision flow is executed based on the insurance product digital model, the claim settlement knowledge base, and the structured claim settlement data to perform claim settlement audit and generate claim settlement audit results, including: Based on the material integrity rules, it is verified whether the structured claim settlement data covers all required elements for claim settlement. If there is a missing, a material completion prompt is generated. Based on the product liability matrix and the structured liability tree, the diagnosis information, treatment plan and insurance liability scope in the structured claim settlement data are matched to verify the indication compliance. A preset disease evolution model is used to analyze the disease course information in the structured claim settlement data to verify the logical consistency of the disease course. Based on the risk interception rules, the structured claim settlement data is risk-identified, and if high risk is identified, the case is marked for manual review. The cases that pass the verification are calculated for the amount of compensation and the audit pass results are generated. The cases that do not pass the verification are generated with the audit failure results containing the reasons for the failure. The claim settlement audit results include the audit pass results and the audit failure results.

6. The method of claim 5, wherein, The preset disease evolution model is used to analyze the disease course information in the structured claim settlement data to verify the logical consistency of the treatment, including: Extracting pathological indicators, diagnosis results, treatment plans and medication records from the structured claim settlement data to construct a medical disease course chain with a time axis; Input the medical disease course chain into the preset disease evolution model to obtain a disease progression prediction trajectory; Compare the matching degree of the actual treatment information in the medical disease course chain with the disease progression prediction trajectory. If the matching degree is lower than a threshold, it is determined that the treatment logic is inconsistent.

7. The method of claim 1, wherein, The claim settlement audit results and claim settlement progress information are fed back to the user through the multi-terminal intelligent claim settlement channel, and the claim settlement operation is performed according to the claim settlement audit results, including: The claim settlement audit results are pushed through the multi-terminal intelligent claim settlement channel. If the audit passes, the compensation related information is synchronized. If the audit fails, the reasons and supplementary requirements are explained. Real-time update of claim settlement progress information, push process node change notification through the multi-terminal intelligent claim settlement channel; Trigger the compensation instruction for the cases that pass the audit and send the account notification. For cases where the user supplements the materials, the supplementary materials are used as new multi-modal claim settlement materials to re-execute intelligent analysis and audit process.

8. An AI-based intelligent insurance claim processing device, characterized in that, It includes: A knowledge base construction module for constructing an insurance product digital model and a claim settlement knowledge base; A claim settlement channel deployment module for deploying a multi-terminal intelligent claim settlement channel to receive user-submitted claim settlement applications and multi-modal claim settlement materials through the multi-terminal intelligent claim settlement channel; A material analysis processing module for analyzing the multi-modal claim settlement materials using an intelligent material processing engine to obtain structured claim settlement data. The analysis processing includes at least one of material classification, image optimization, key information identification, and medical text semantic analysis. An audit result determination module for performing claim settlement audit based on the insurance product digital model, the claim settlement knowledge base, and the structured claim settlement data through an AI intelligent audit decision flow to generate claim settlement audit results. The claim settlement result execution module is configured to feed back the claim settlement audit result and claim settlement progress information to the user through the multi-end intelligent claim settlement channel and execute claim settlement operation according to the claim settlement audit result.

9. An electronic device, comprising: The electronic device includes: at least one processor; and a memory connected in communication with 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 to enable the at least one processor to execute the AI-based intelligent insurance claim settlement processing method of any one of claims 1-7.

10. A computer-readable storage medium, characterized in that, The computer readable storage medium stores computer instructions for causing the processor to implement the AI-based intelligent insurance claim settlement processing method of any one of claims 1-7 when executed.

Images