Intelligent claim settlement method and device, electronic device, and storage medium

Abstract

The intelligent claim settlement method and device, the electronic equipment and the storage medium provided by the present application are related to the technical field of artificial intelligence and are suitable for the fields of finance and medical health. The method comprises the following steps: acquiring multi-modal claim settlement application data; performing disease correlation on the multi-modal claim settlement application data based on a preset knowledge graph to obtain a target disease; performing health assessment on a target object according to the multi-modal claim settlement application data and the target disease to obtain disease severity; performing treatment evaluation according to the multi-modal claim settlement application data and a target symptom to obtain treatment rationality; performing cost evaluation according to the multi-modal claim settlement application data and a target treatment scheme to obtain cost relevance; performing data fusion according to the disease severity, the treatment rationality and the cost relevance to obtain a claim settlement suggestion; and performing claim settlement on a claim settlement application according to the target disease and the claim settlement suggestion. The present application can improve the accuracy and efficiency of claim settlement for insurance.

Description

Technical Field

[0001] This application relates to the field of artificial intelligence technology, and is applicable to the financial and healthcare fields, particularly to an intelligent claims processing method and device, electronic device, and storage medium. Background Technology

[0002] In the field of artificial intelligence technology, pre-trained large models or intelligent agents can be used to provide automated claims services for users applying for insurance claims.

[0003] Currently, insurance companies and other financial institutions face the following shortcomings when processing claims for insurance (such as health insurance): (1) The relevant technologies struggle to efficiently integrate multimodal medical data such as electronic medical records, image data, and gene testing, making it impossible to accurately determine whether a patient's disease is eligible for reimbursement. (2) Existing AI models rely on historical data for training, resulting in a significant decrease in the accuracy of identifying diseases with low incidence rates, which can easily lead to misdiagnosis or missed diagnosis, requiring manual intervention. Therefore, the relevant technologies cannot effectively identify diseases, easily causing interruptions in claims services. Summary of the Invention

[0004] The main objective of this application is to propose intelligent claims processing methods, devices, equipment, and media that can solve the technical problem of high difficulty in disease identification, improve the accuracy of disease identification, and thus improve the accuracy and efficiency of insurance claims processing.

[0005] To achieve the above objectives, a first aspect of this application proposes an intelligent claims settlement method, the method comprising: In response to the detection of a claim application from a target object, multimodal claim application data is obtained; Disease association is performed on the multimodal claims data based on a preset knowledge graph to obtain the target disease; wherein, the preset knowledge graph is constructed based on the relationship between preset diseases, disease symptoms, and treatment plans; Based on the multimodal claims data and the target disease, a health assessment is performed on the target individual to obtain the disease severity. Obtain the target symptoms of the target disease, and conduct a treatment assessment based on the multimodal claim data and the target symptoms to determine the rationality of the treatment; Obtain the target treatment plan for the target disease, and conduct a cost assessment based on the multimodal claim data and the target treatment plan to obtain the cost correlation. Based on the severity of the disease, the reasonableness of the treatment, and the relevance of the costs, a claims assessment recommendation is obtained. The claim will be processed based on the target disease and the claim assessment recommendation.

[0006] Optionally, the method is applied to a claims system, which includes a managed claims platform, a digital intelligence platform, and a data middleware platform; the step of obtaining multimodal claims application data in response to detecting a claims application from a target object includes: In response to the detection of a claim application from a target object, the insurance case of the target object is determined through the managed claims platform, and the case identifier of the insurance case is transmitted to the digital platform; The system receives the case identifier through the digital intelligence platform and sends a case data request to the data platform based on the case identifier. Through the data platform, data is read from at least two data sources based on the case data request to obtain multimodal claim application data, and the multimodal claim application data is transmitted to the digital intelligence platform.

[0007] Optionally, the claims system further includes an intelligent diagnostic engine, which is bound to the preset knowledge graph; The process of associating the multimodal claims data with diseases based on a preset knowledge graph to obtain target diseases includes: The digital intelligence platform generates a disease reasoning task based on the multimodal claims application data and transmits the disease reasoning task to the intelligent diagnostic engine. The intelligent diagnostic engine matches the target disease with the preset knowledge graph and the multimodal claims application data.

[0008] Optionally, the step of matching the target disease based on the preset knowledge graph and the multimodal claims application data includes: The multimodal claims data is fused using a large language model to obtain disease description text; Based on the disease description text, the disease symptoms and treatment plans of candidate diseases in the preset knowledge graph are matched to obtain the matching confidence of the candidate diseases; The candidate diseases are screened based on the matching confidence level to obtain the target disease.

[0009] Optionally, before performing disease association on the multimodal claims application data based on a preset knowledge graph to obtain the target disease, the method further includes: Obtain raw medical data, including raw disease data, raw symptom data, and raw treatment plan data; Entity extraction is performed based on the original disease data to obtain disease nodes, entity extraction is performed based on the original symptom data to obtain symptom nodes, and entity extraction is performed based on the original treatment plan data to obtain treatment plan nodes. Create a first edge between the disease node and the symptom node, create a second edge between the disease node and the treatment plan, and create a third edge between the symptom node and the treatment plan; Using medical literature evidence, attention is calculated between the disease node and the symptom node to obtain the first association weight of the first side; attention is calculated between the disease node and the treatment plan node to obtain the second association weight of the second side; and attention is calculated between the symptom node and the treatment plan node to obtain the third association weight of the third side, so as to construct the preset knowledge graph. The preset knowledge graph is bound to the intelligent diagnostic engine.

[0010] Optionally, the claims system further includes a multi-source monitoring agent platform. After binding the preset knowledge graph to the intelligent diagnostic engine, the method further includes: The multi-source monitoring agent platform detects the medical database, and when an update is detected in the medical database, it acquires the new medical data and transmits the new medical data to the intelligent diagnostic engine. The intelligent diagnostic engine updates the preset knowledge graph based on the newly added medical data.

[0011] Optionally, the claims system further includes a claims review rule engine, wherein the step of performing a health assessment on the target object based on the multimodal claims application data and the target disease to obtain the disease severity includes: The claims review rule engine extracts pathology reports and imaging reports from the multimodal claims application data. The claims processing rule engine is used to assess disease progression based on the pathology report, the imaging report, and the target disease, thereby obtaining the disease progression level of the target individual. The treatment tolerance score of the target subject is obtained by using the claims review rule engine based on the pathology report, the imaging report and the target disease to conduct a treatment assessment. The disease severity is obtained by fusing the disease progression level and the treatment tolerance score.

[0012] To achieve the above objectives, a second aspect of this application provides an intelligent claims processing device, the device comprising: The data acquisition module is used to acquire multimodal claim application data in response to the detection of a claim application from a target object; The disease association module is used to associate diseases in the multimodal claims application data based on a preset knowledge graph to obtain the target disease; wherein, the preset knowledge graph is constructed based on the relationship between preset diseases, disease symptoms, and treatment plans; The health assessment module is used to perform a health assessment on the target object based on the multimodal claim application data and the target disease, and to obtain the severity of the disease; The treatment assessment module is used to obtain the target symptoms of the target disease, and to conduct a treatment assessment based on the multimodal claim data and the target symptoms to determine the rationality of the treatment. The cost assessment module is used to obtain the target treatment plan for the target disease, and to assess the cost based on the multimodal claim data and the target treatment plan to obtain the cost correlation. The insurance claims module is used to fuse data based on the severity of the disease, the reasonableness of the treatment, and the relevance of the cost to obtain claims recommendations. The insurance claims module is used to process the claims application based on the target disease and the claims review recommendation.

[0013] To achieve the above objectives, a third aspect of this application provides an electronic device, which includes a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the intelligent claims method described in the first aspect.

[0014] To achieve the above objectives, a fourth aspect of the present application provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the intelligent claims method described in the first aspect.

[0015] The intelligent claims processing method, intelligent claims processing device, electronic device, and computer-readable storage medium proposed in this application first respond to the detection of a claims application from a target object by acquiring multimodal claims application data. Then, using a pre-defined knowledge graph carrying relationships between pre-defined diseases, disease symptoms, and treatment plans, the multimodal claims application data is correlated with diseases to obtain the target disease. After obtaining the target disease, health assessment, treatment assessment, and cost assessment can be performed in parallel to ultimately generate a claims settlement recommendation. Finally, the claims application is processed based on the target disease and the claims settlement recommendation. In this way, this application solves the technical problem of high difficulty in disease identification, improves the accuracy of disease identification, and thus improves the accuracy and efficiency of insurance claims processing.

[0016] Additional aspects and advantages of this application will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of this application. Attached Figure Description

[0017] Figure 1 This is a flowchart of the intelligent claims settlement method provided in the embodiments of this application; Figure 2 yes Figure 1The flowchart for step 101 in the document; Figure 3 yes Figure 1 The flowchart for step 102 in the document; Figure 4 yes Figure 3 The flowchart for step 302 in the document; Figure 5 yes Figure 1 The flowchart for step 103 in the text; Figure 6 This is a block diagram of the module structure of the intelligent claims processing device provided in the embodiments of this application; Figure 7 This is a schematic diagram of the hardware structure of the electronic device provided in the embodiments of this application. Detailed Implementation

[0018] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0019] It should be noted that although functional modules are divided in the device schematic diagram and a logical order is shown in the flowchart, in some cases, the steps shown or described may be performed in a different order than the module division in the device or the order in the flowchart. The terms "first," "second," etc., in the specification, claims, and the aforementioned drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence.

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

[0021] First, let's analyze some of the terms used in this application: Artificial intelligence (AI) is a new branch of computer science that studies, develops, and applies theories, methods, technologies, and systems to simulate, extend, and expand human intelligence. It aims to understand the essence of intelligence and produce intelligent machines that can react in a way similar to human intelligence. Research in this field includes robotics, speech recognition, image recognition, natural language processing, and expert systems. AI can simulate the information processes of human consciousness and thought. Furthermore, AI utilizes digital computers or machines controlled by digital computers to simulate, extend, and expand human intelligence, perceiving the environment, acquiring knowledge, and using that knowledge to achieve optimal results.

[0022] Natural Language Processing (NLP): NLP uses computers to process, understand, and utilize human language (such as Chinese and English). It is a branch of artificial intelligence and an interdisciplinary field of computer science and linguistics, often referred to as computational linguistics. NLP includes syntactic analysis, semantic analysis, and discourse understanding. It is commonly used in machine translation, handwritten and printed character recognition, speech recognition and text-to-speech conversion, information and image processing, information extraction and filtering, text classification and clustering, sentiment analysis, and opinion mining. It involves data mining, machine learning, knowledge acquisition, knowledge engineering, artificial intelligence research, and linguistic research related to language computation.

[0023] An AI Agent is an application that can autonomously plan and invoke external tools to execute tasks based on objectives and external information, and reflect on the results. AI Agents are designed around task objectives and business processes in terms of implementation, operational form, and behavioral characteristics.

[0024] Large Language Models (LLMs) are a class of deep learning models that utilize massive amounts of text data for autoregressive or self-supervised learning. They are capable of generating, understanding, translating, and summarizing complex natural language tasks. Common underlying frameworks are often based on the Transformer architecture, with core elements including attention mechanisms, multi-layer stacking, and positional encoding. Typical capabilities include text generation, question answering, dialogue, summarization, translation, code completion, and sentiment analysis.

[0025] A knowledge graph (KG) is a graph-based method for representing and managing knowledge, where "entities" (nodes) are connected by "relationships" (edges) to form a queryable and reasonable knowledge network. Entities are typically real-world things or concepts (such as people, places, companies, products, events, etc.), while relationships describe the semantic connections between entities (such as "belongs to," "creator," "located in," "same category," etc.). The goal of a knowledge graph is to integrate structured, semi-structured, and unstructured data into a unified, reasonable knowledge resource, supporting semantic queries, reasoning, and the discovery of new relationships.

[0026] Overview of existing similar products and technologies: (1) Automation and intelligence of claims process through robotics, but its core technology focuses on process optimization and lacks in-depth support for critical illness diagnosis. (2) Virtual health assistants provide online consultation and insurance plan recommendations based on user medical history, but do not deeply integrate claims and diagnosis, and their functions are limited to information interaction. (3) AI-assisted precision diagnosis tools: Some commercial insurance companies have tried to use AI for disease risk prediction, such as using machine learning to analyze historical data to identify potential health risks, but most of them remain at the data modeling stage and have not formed a closed-loop "diagnosis-claims" linkage mechanism. (4) Claims system: AI large model technology is used to improve claims efficiency, but it mainly focuses on the automation of document review and does not cover intelligent diagnosis support for critical illness cases. (5) Automation of insurance information entry through OCR and other technologies, but lacks the ability to assist in the diagnosis of complex diseases. (6) Medical resource coordination and expert docking are provided, but they rely on manual intervention and have limited intelligence.

[0027] Existing technical defects and deficiencies: (1) Data integration and standardization issues: (1.1) Poor compatibility of multi-source heterogeneous data: Existing systems have difficulty efficiently integrating multimodal medical information such as electronic medical records, image data, and gene testing, resulting in limited input to diagnostic models. (1.2) Insufficient unstructured data processing capabilities: The semantic parsing accuracy of medical texts (such as doctors' handwritten medical records) is low, affecting the reliability of intelligent diagnosis. (2) Algorithm limitations: (2.1) Weak diagnostic capabilities for rare diseases: Existing AI models rely on historical data for training, resulting in a significant decrease in the accuracy of identifying serious diseases with low incidence rates, which can easily lead to misdiagnosis or missed diagnosis. (2.2) Lack of dynamic risk prediction: Most systems only assess risk based on static data and lack the ability to monitor and dynamically adjust the patient's disease progression in real time.

[0028] (3) Insufficient process coordination: (3.1) The diagnosis and claims process is fragmented: Existing technologies focus on a single process (such as claims automation or disease screening) and have not achieved intelligent coordination of the entire chain of "diagnosis-underwriting-claims". (3.2) Low efficiency of cross-system connection: The interface standards between hospital HIS system, insurance company core system and third-party health management platform are not unified, and there are barriers to data flow.

[0029] (4) User experience pain points: (4.1) Insufficient user-friendliness: Existing tools mostly use professional terminology interfaces, which makes it difficult for ordinary users to operate and results in low self-service usage. (4.2) Service response delay: Complex serious illness cases still require manual review, with an average processing cycle of more than 48 hours, which cannot meet the needs of emergency patients.

[0030] Based on this, the embodiments of this application propose an intelligent claims method, intelligent claims device, electronic device, and computer-readable storage medium, which can solve the core problems existing in the current claims system, such as data fragmentation, disconnect between diagnosis and claims process, and low efficiency of tag reference.

[0031] The intelligent claims settlement method provided in this application can be applied to terminals and servers, or it can be software running on the server. The server can be configured as an independent physical server, or as a server cluster or distributed system composed of multiple physical servers, or it can be configured as a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, CDN, and big data and artificial intelligence platforms; the software can be an application that implements the intelligent claims settlement method, etc., but is not limited to the above forms.

[0032] This application can be used in a wide variety of general-purpose or special-purpose computer system environments or configurations. Examples include server computers, multiprocessor systems, microprocessor-based systems, set-top boxes, programmable consumer electronics devices, network PCs, and distributed computing environments including any of the above systems or devices. This application can be described in the general context of computer-executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc., that perform specific tasks or implement specific abstract data types. This application can also be practiced in distributed computing environments where tasks are performed by remote processing devices connected via a communication network. In distributed computing environments, program modules can reside in local and remote computer storage media, including storage devices.

[0033] This application provides an intelligent claims settlement method, an intelligent claims settlement device, an electronic device, and a computer-readable storage medium, which are specifically described through the following embodiments. First, the intelligent claims settlement method in the embodiments of this application is described.

[0034] It should be noted that in each specific implementation of this application, when it is necessary to process data related to the user's identity or characteristics, such as the user's personal health information (electronic medical records, image reports), the user's permission or consent will be obtained first. Moreover, the collection, use and processing of this data will comply with relevant laws, regulations and standards.

[0035] Reference Figure 1 , Figure 1 This is an optional flowchart of the intelligent claims settlement method provided in this application embodiment, which can be applied to a claims settlement system. The method may include, but is not limited to, steps 101 to 107.

[0036] Step 101: In response to detecting a claim application from the target object, obtain multimodal claim application data; Step 102: Based on a preset knowledge graph, perform disease association on the multimodal claims application data to obtain the target disease; wherein, the preset knowledge graph is constructed based on the relationship between preset diseases, disease symptoms, and treatment plans; Step 103: Conduct a health assessment of the target individuals based on multimodal claim application data and the target disease to obtain the severity of the disease; Step 104: Obtain the target symptoms of the target disease, and conduct a treatment assessment based on the multimodal claim data and target symptoms to determine the rationality of the treatment; Step 105: Obtain the target treatment plan for the target disease, and conduct a cost assessment based on the multimodal claim data and the target treatment plan to obtain the cost correlation. Step 106: Data fusion is performed based on disease severity, treatment rationality, and cost relevance to obtain claims assessment recommendations; Step 107: Process the claim based on the target disease and the claim assessment recommendations.

[0037] Steps 101 to 107, as illustrated in this embodiment, firstly, in response to the detection of a claim application from a target object, multimodal claim application data is acquired. Then, using a preset knowledge graph containing relationships between preset diseases, disease symptoms, and treatment plans, disease association is performed on the multimodal claim application data to obtain the target disease. After obtaining the target disease, health assessment, treatment assessment, and cost assessment can be performed in parallel to ultimately generate a claim settlement recommendation. Finally, the claim application is processed based on the target disease and the claim settlement recommendation. In this way, this application can solve the technical problem of high difficulty in disease identification, improve the accuracy of disease identification, and thus improve the accuracy and efficiency of insurance claims processing.

[0038] For example, in health insurance scenarios within the financial and healthcare sectors, multimodal claims data can include electronic medical records, imaging reports, wearable device monitoring data (such as heart rate and blood oxygen saturation), and genetic testing results (VCF format). Based on this multimodal claims data, it can be identified that the target individual is suffering from metastatic lung cancer, and the claims assessment recommendation includes whether to pay, the recommended payout amount, and the scope and items covered. Finally, the health insurance claim is processed based on the target disease being metastatic lung cancer and the claims assessment recommendation.

[0039] In step 101 of some embodiments, in response to detecting a claim application from a target object, multimodal claim application data is obtained. Target object: refers to the insured or their agent who submits the claim. Claim application: refers to a request submitted by the insured, beneficiary, or their agent to the insurance company to obtain compensation to cover compliant costs incurred due to health-related services such as illness, medical expenses, hospitalization, outpatient treatment, and rehabilitation. Multimodal claim application data: refers to claim application materials containing multiple modalities of information, such as text descriptions, images / videos, audio / video recordings, structured forms, sensor data, etc., used to comprehensively assess the claim situation. For example, multimodal claim application data includes electronic medical records, imaging reports (DICOM), wearable device monitoring data (such as heart rate, blood oxygen), and genetic testing results (VCF format).

[0040] In one example, the insured submits a health insurance claim to the claims system for complications related to hypertension. At this time, multimodal claim application data can be obtained from the claims system, which may include: text such as claim application form, medical history description, chief complaint, etc.; images such as scanned copies of hospital records, prescription images, and digitized files of CT / MRI images; structured data such as admission date, diagnosis code (e.g., ICD-10 / WHO), medication list, length of hospital stay, and confirmed medical record number; and audio and video such as transcripts of doctor's outpatient explanation videos and time series of ward monitoring data (e.g., blood pressure, heart rate, etc.).

[0041] In one example, an adaptive data cleaning framework can be used to unify the data format of multimodal claims application data. Furthermore, NLP technology can be used to perform entity recognition (such as disease names and medication dosages) on unstructured data such as handwritten medical records, combined with OCR to automatically extract key fields from image reports (such as lesion size annotations in CT images) to obtain structured data.

[0042] In one embodiment, the claims system includes a managed claims platform, a digital intelligence platform, a data middleware, an intelligent diagnostic engine, and a claims settlement rule engine.

[0043] The main responsibilities of a managed claims platform include: Full-process claims management: closed-loop management from case filing, document collection, preliminary review, loss assessment, claims settlement, claims payment to case closure and archiving. Workflow and task scheduling: allocating tasks according to rules, timeliness, and personnel responsibilities to ensure timely completion of each step. Approval and access control: multi-level approval, role separation, access control, and audit record retention. The core functions of a managed claims platform include: Case management: case creation, status tracking, timeline display, and instant reminders. Document management and version control: document uploading, alignment, version history, and missing document prompts. Rule-driven claims settlement path: integrating claims settlement rules to guide cases towards automated judgment or manual review. Evidence management and evidence chain traceability: traceability of evidence sources, evidence levels, and evidence mapping relationships. Statistics and reports: claims settlement timeliness, case closure rate, claims classification, and abnormal case analysis. The managed claims platform shares structured data and unstructured documents with the digital intelligence platform and data middleware. The managed claims platform pushes pending claims tasks and evidence summaries to the intelligent diagnostic engine and the claims review rules engine.

[0044] The main responsibilities of the digital intelligence platform include: data governance and data asset management: data standardization, data lineage, metadata management, and data quality monitoring. Unified entry point and service orchestration: providing unified service orchestration, API gateway, and authentication and authorization for upper-layer applications. Support for driving analytics and intelligent applications: providing data analysis, visualization, and model service registration and invocation.

[0045] The core functions of the digital intelligence platform include: Data Lake / Data Warehouse: storing structured and unstructured data, supporting large-scale querying and analysis. Data Cleaning and Quality Control: data cleaning, deduplication, missing value handling, outlier detection, and data consistency checks.

[0046] The main responsibilities of a data platform include: Data standardization and semantic unification: establishing a unified data model and coding system. Data integration and interoperability: connecting with external systems (EHR, imaging systems, payment systems, external commercial insurance data, etc.) to achieve data aggregation. Data security and compliance control: permissions, data masking, data classification, auditing, and compliance reporting. The core functions of a data platform include: Unified data model: defining core entities and relationships such as policies, claims, patients, and medical evidence. Data governance and quality system: quality indicators, data cleaning rules, metadata catalog, and data lineage. Data service layer: CRUD-style APIs, data access strategies, caching, and engine optimization.

[0047] The main function of the intelligent diagnostic engine is to output the target disease based on input data (including a preset knowledge graph and multimodal claim application data). The main function of the claims review rule engine is to output claims review suggestions based on input data (including the target disease, multimodal claim application data, and the terms and conditions library).

[0048] In one embodiment, reference is made to Figure 2 Step 101 may include: Step 201: In response to the detection of a claim application from the target object, the insurance case of the target object is determined through the managed claims platform, and the case identifier of the insurance case is transmitted to the digital platform; Step 202: Receive the case identifier through the digital intelligence platform and send a case data request to the data platform based on the case identifier; Step 203: Through the data platform, data is read from at least two data sources based on the case data request to obtain multimodal claim application data, and then the multimodal claim application data is transmitted to the digital intelligence platform.

[0049] Data sources can include medical data sources (such as electronic medical record systems, image databases, and wearable device testing data) and claims data sources (such as insurance information databases and customer information management systems).

[0050] The advantage of the embodiments of steps 201 to 203 above is that, through the interaction between the managed claims platform, the digital intelligence platform and the data platform, multimodal claims application data can be obtained, and the multimodal claims application data will not enter the managed claims platform, thereby reducing the risk of data leakage and ensuring data security.

[0051] In step 102 of some embodiments, disease association is performed on multimodal claims application data based on a preset knowledge graph to obtain the target disease. The preset knowledge graph is a pre-constructed set of nodes and edges. Nodes represent preset diseases, disease symptoms, and treatment plans, and edges represent the relationships between nodes. The preset knowledge graph is constructed based on the relationships between preset diseases, disease symptoms, and treatment plans. For example, the preset knowledge graph includes: disease nodes: cerebral infarction, cerebral hemorrhage, hypertension, etc.; symptom nodes: headache, limb weakness, speech impairment, etc.; treatment plan nodes: antiplatelet drugs, antihypertensive drugs, thrombolytic therapy, etc. The target disease is the specific disease that is of primary concern in the current claims scenario.

[0052] In one embodiment, a preset knowledge graph and multimodal claim application data can be input into a large language model, which will then select at least one preset disease from the preset knowledge graph as the target disease.

[0053] In one example, if the multimodal claims data indicates that the patient has headaches, blurred vision, a history of hypertension, and imaging shows brain abnormalities, then the target disease is identified based on the "disease-symptom-treatment" relationship in the preset knowledge graph. For example, if "headache" is strongly associated with "cerebrovascular disease" and the imaging evidence points to vascular lesions, then cerebrovascular disease is listed as one of the target diseases.

[0054] In one embodiment, the intelligent diagnostic engine is bound to a preset knowledge graph; refer to Figure 3 Step 102 may include: Step 301: Generate a disease reasoning task based on multimodal claims application data through the digital intelligence platform, and send the disease reasoning task to the intelligent diagnostic engine; Step 302: Using an intelligent diagnostic engine, the target disease is obtained by matching the disease with a preset knowledge graph and multimodal claim application data.

[0055] Specifically, although the digital intelligence platform can obtain multimodal claims application data from the data middle platform, in order to further achieve data isolation, this embodiment uses an intelligent diagnostic engine for disease matching, which further improves data security. Moreover, when errors occur in disease matching, the intelligent diagnostic engine can be quickly maintained / updated, avoiding frequent maintenance of the digital intelligence platform and facilitating the normal operation of the claims system.

[0056] In one embodiment, reference is made to Figure 4 Step 302 may include: Step 401: Multimodal data fusion is performed on the multimodal claims application data using a large language model to obtain disease description text; Step 402: Match the disease symptoms and treatment plans of candidate diseases in the preset knowledge graph based on the disease description text to obtain the matching confidence of the candidate diseases; Step 403: Filter candidate diseases based on matching confidence to obtain the target disease.

[0057] In step 401, the disease description text represents the target patient's disease symptoms and treatment plan. For example, the disease description text is: The patient is a 65-year-old male who complains of persistent cough with increased sputum. In the past 3 months, he has experienced shortness of breath, decreased exercise tolerance, and worsening nocturnal dyspnea and chest tightness. He is currently using a long-acting bronchodilator daily to improve airway narrowing, with short-acting rescue medications added as needed.

[0058] In step 402, the similarity between the disease symptoms and treatment plans of candidate diseases in the preset knowledge graph and the disease description text can be calculated, and the calculated similarity can be used as the matching confidence.

[0059] In step 403, the higher the matching confidence, the greater the probability that the candidate disease will be the target disease.

[0060] The advantage of the embodiments of steps 401 to 403 above is that they cleverly transform multimodal claim application data into disease description text, thereby finding the target disease from the preset knowledge graph, reducing the difficulty of disease identification and improving the accuracy of disease identification.

[0061] In one embodiment, prior to step 302, the intelligent claims method further includes: constructing a graph based on the relationship between preset diseases, disease symptoms, and treatment plans to obtain a preset knowledge graph, specifically including: Obtain raw medical data, which includes raw disease data, raw symptom data, and raw treatment plan data; Entity extraction is performed based on the original disease data to obtain disease nodes, entity extraction is performed based on the original symptom data to obtain symptom nodes, and entity extraction is performed based on the original treatment plan data to obtain treatment plan nodes. Create the first edge between the disease node and the symptom node, create the second edge between the disease node and the treatment plan, and create the third edge between the symptom node and the treatment plan; Using medical literature evidence, attention is calculated on disease nodes and symptom nodes to obtain the first association weight of the first side, attention is calculated on disease nodes and treatment plan nodes to obtain the second association weight of the second side, and attention is calculated on symptom nodes and treatment plan nodes to obtain the third association weight of the third side, so as to construct the preset knowledge graph. Bind the pre-set knowledge graph to the intelligent diagnostic engine.

[0062] Specifically, when constructing the pre-defined knowledge graph, (1) based on the ICD-10 coding system, a unified disease classification standard is established, symptoms are finely coded, and treatment plans (such as drugs) are classified. By mapping the three types of information—disease, disease symptoms, and treatment plans—to the standardized coding system, a three-dimensional association foundation with a clear structure and explicit semantics is constructed, achieving semantic alignment across data sources. (2) Based on the automatic association construction of graph neural networks, unlike the traditional method that relies on experts to manually input rules, this embodiment utilizes graph neural networks and attention mechanisms to automatically learn the complex association relationships between diseases, disease symptoms, and treatment plans from a large amount of real data, and dynamically calculates the association weights among the three. This allows the graph to not only reflect general laws but also capture subtle differences in clinical practice, improving the accuracy of intelligent judgment.

[0063] In one embodiment, the claims system further includes a multi-source monitoring agent platform. After the step of binding the preset knowledge graph to the intelligent diagnostic engine, the intelligent claims method further includes: detecting the medical database through the multi-source monitoring agent platform, obtaining new medical data when an update is detected in the medical database, and transmitting the new medical data to the intelligent diagnostic engine; and updating the preset knowledge graph based on the new medical data through the intelligent diagnostic engine.

[0064] Specifically, the pre-defined knowledge graph has real-time update capabilities. When new medical data is added (such as changes in treatment standards), the claims system can automatically identify and trigger a local graph update. The interactive claims management platform provides a visual editing interface, allowing business operations to adjust tag associations and record all modification paths, enabling continuous optimization and traceable management of the knowledge graph.

[0065] In step 103 of some embodiments, a health assessment is performed on the target individual based on multimodal claims data and the target disease to obtain the disease severity. Health assessment: A comprehensive evaluation of an individual's health status, disease state, complications, etc., typically including a comprehensive judgment of information such as physical signs, laboratory indicators, and imaging. Disease severity: Refers to a graded score of the severity of a disease. Disease severity is a classification of the risk and progression of the disease (e.g., A / B / C level, or mild / moderate / severe, or a 1-5 scale), reflecting the magnitude of the health risk.

[0066] In one example, if the target disease is ischemic stroke, the severity of the target disease is calculated by combining multimodal claims data, including imaging reports (such as the size and extent of the infarct area on brain CT / MRI), neurological function scores (such as the NIHSS score), and biochemical indicators (blood glucose, blood lipids, inflammatory markers). For instance, using the NIHSS score as the primary indicator, an NIHSS score ≥ 15 is considered a severe stroke; an NIHSS score between 5 and 14 is considered moderate; and ≤ 4 is considered mild.

[0067] In one embodiment, the claims system includes a claims review rules engine, which is bound to a terms and conditions library. This terms and conditions library is a structured rule base based on ICD-10 classification rules, NMPA indication restrictions, and the CMDE treatment catalog. For example, the terms and conditions library includes the following fields: Terms ID: Unique identifier, such as TL-ICD10-001; Version / Effective Date; Source Rule Domain: ICD10 classification rule, or NMPA indication restrictions, or CMDE treatment catalog; Applicable Insurance Type / Scenario: Health Insurance - Inpatient Treatment / Outpatient Treatment; Liability Category: Treatment Liability, Payment Limit, Deductible / Deductible Amount, Drug Compatibility; Terminology and Coding: Disease Code (ICD-10), Treatment Item Code, Drug / Treatment Code; Rule Highlights: Summary Description; Evidence and Evidence Level Requirements: Medical Orders, Diagnosis Certificates, Inpatient Medical Records, Imaging, Examination Reports; Calculation Scope / Payment Scope: Reimbursement Ratio, Deductible, Maximum Limit, Depreciation, Medical Insurance Settlement, etc.; Exceptions and Conflict Handling: Terms Priority, Conflict Resolution Rules; Audit and Compliance Highlights: Audit Path, Change Records; Citations / Links: Original Terms and Conditions, Regulatory Terms, Guideline Links; Evidence Mapping Table: List of Required Materials; Confidence Level / Judgment Result Format: Applicability Score, Confidence Interval. The rules engine can perform health assessments, treatment assessments, and cost assessments in parallel, and then combine them with the terms library to generate claims recommendations.

[0068] In one embodiment, reference is made to Figure 5 Step 103 may include: Step 501: Extract pathology reports and imaging reports from multimodal claims application data using the claims review rule engine; Step 502: Using the claims rule engine, assess the disease progression based on the pathology report, imaging report, and target disease to obtain the disease progression level of the target individual; Step 503: Using the claims rule engine, conduct a treatment assessment based on the pathology report, imaging report, and target disease to obtain the treatment tolerance score of the target subject; Step 504: Combine the disease progression level and treatment tolerance score to obtain the disease severity.

[0069] In step 501, the pathology report includes pathological diagnosis, classification, staging, landmark pathological features, tumor size, histological grade, etc. The imaging report refers to imaging results such as CT / MRI / PET-CT, including the number and size of lesions, degree of enhancement, metastasis, imaging staging, etc.

[0070] In step 502, the claims processing rule engine can invoke a disease progression assessment model (such as a neural network model trained on supervised medical data) to evaluate the pathology report, imaging report, and target disease to obtain the disease progression level of the target object. For example, if the lesion diameter increases by more than a certain percentage and the imaging indicates a new lesion, the disease progression level is upgraded to moderate or severe progression. If the pathology shows a significant change in malignant characteristics consistent with the imaging, the disease progression level is upgraded to a high level.

[0071] In step 503, the claims processing rule engine invokes the treatment assessment model to evaluate the pathology report, imaging report, and target disease, obtaining a treatment tolerance score for the target patient. The treatment tolerance score is a quantitative, score-based assessment of a patient's tolerance to medications, treatment methods, and treatment intensity under a specific treatment regimen. For example, pathology and imaging reports can reflect the incidence, severity, and recovery status of abnormalities in the target patient's blood count, liver and kidney function, and inflammatory markers; the claims processing rule engine then outputs the treatment tolerance score based on this information.

[0072] In step 504, the higher the disease progression grade, the higher the disease severity; the higher the treatment tolerance score, the lower the disease severity.

[0073] The benefit of the embodiments of steps 501 to 504 described above is that by introducing disease progression grade and treatment tolerance score to comprehensively assess disease severity, the accuracy of health assessment is improved, which helps to improve the accuracy of claims processing.

[0074] In step 104 of some embodiments, the target symptoms of the target disease are obtained. The claims rule engine performs a treatment assessment based on multimodal claims application data and the target symptoms to determine the rationality of the treatment. Target symptoms: Clinical manifestations or signs directly related to the target disease that require attention (such as fever, pain, difficulty breathing, etc.). Target symptoms can be obtained from preset knowledge graphs, treatment guidelines, etc. Treatment assessment: Judging the rationality, feasibility, and degree of evidence support of the proposed treatment measures (medication, surgery, physical therapy, etc.). In one example, the target symptom is ischemic stroke with hemiplegia. The claims rule engine assesses whether the treatment is timely and compliant based on the target symptoms (hemiplegia, speech impairment), treatment plan (intravenous thrombolysis, secondary prevention drugs, rehabilitation therapy), and treatment guidelines.

[0075] In step 105 of some embodiments, a target treatment plan for the target disease is obtained. The claims rule engine performs a cost assessment based on multimodal claims data and the target treatment plan to obtain cost relevance. Target treatment plan: A specific combination of treatments proposed to achieve the diagnostic and treatment goals. Target treatment plans can be obtained from preset knowledge graphs, treatment guidelines, etc. Cost relevance: The degree of consistency, necessity, and reasonableness between the treatment plan and the actual costs incurred. In one example, the treatment plan for an ischemic stroke patient includes thrombolytic drugs, antiplatelet drugs, and rehabilitation therapy. The claims rule engine combines drug prices, length of hospital stay, rehabilitation training duration, and equipment usage to assess the consistency between the cost of thrombolytic drugs and actual expenditures, and whether there are unreasonable additional charges (such as duplicate drugs, costs not included in the medical insurance catalog, etc.).

[0076] In step 106 of some embodiments, data fusion is performed based on disease severity, treatment rationality, and cost relevance to obtain a claims assessment recommendation. Data fusion: Integrating data from different modalities and sources within the same framework to form a unified assessment result or decision-making basis. Claims assessment recommendation: A recommendation on claims decisions and payout amounts based on data analysis and rule-based reasoning, which may include the payout ratio, whether full payout is possible, and the need for supplementary materials.

[0077] In one example, when the target disease is metastatic lung cancer, the claims processing engine generates the following three-dimensional assessment matrix, which includes disease severity, treatment rationality, and cost relevance:

[0078] For example, the three-dimensional assessment matrix is: {Disease Severity: {TNM Staging: pT2N1M0, Overall Score: 3.8, Grade: B}, Treatment Rationality: {Targeted Therapy: Rational, Chemotherapy Regimen: Warning (Dosage Exceeded)}, Cost Relevance: {Genetic Testing: Rational, MRI Examination: Warning (Not Recommended by Guidelines)}. After obtaining the three-dimensional matrix, the claims processing rule engine automatically calls the terms library to output claims processing suggestions. For example, claims processing suggestions include: payment conclusion (e.g., "Agree to full payment / Agree to partial payment / Reject claim / Pending manual review"); payment amount and details: amount payable, deductible deduction, itemized amounts for various fees, taxes, etc.

[0079] In step 107 of some embodiments, the claim application is processed based on the target disease and the claim settlement recommendation. Claim settlement: After an insured event occurs, the insured or beneficiary submits a claim to the insurance company, and the process includes its review, loss assessment, and payment. In one example, the claims system outputs a claim settlement recommendation such as "partial payment, 55% reimbursement ratio, rehabilitation expense invoice required," etc. The claims system maps the claim settlement recommendation to the actual claims process, verifies the deductible, reimbursement ratio, and maximum limit in the policy terms, completes the actual payment, and records the payment voucher and claims result.

[0080] In summary, this application achieves at least the following beneficial effects: Increased efficiency: By integrating multimodal data (electronic medical records, image reports, wearable device detection data, etc.) and combining them with knowledge graphs, the diagnosis time for serious illnesses is reduced from the traditional 4-6 hours to 15 minutes, with automated processing covering 90% of common cases. Industry-leading accuracy: A three-dimensional disease labeling system built based on the ICD-10 standard expands the types of rare diseases that can be identified and improves diagnostic accuracy. Enhanced compliance and risk control: Blockchain-based evidence storage technology enables full traceability of treatment decisions. Significantly optimized costs: It reduces both manual review costs and dispute resolution costs, and data cleaning is fully automated. Closed-loop service ecosystem: By linking community health management support centers and special drug services, it provides one-stop management from diagnosis to rehabilitation, adapts to resources from multiple community chronic disease centers, and further expands the scope of innovative special drug claims.

[0081] Please see Figure 6 This application also provides an intelligent claims processing device that can implement the above-mentioned intelligent claims processing method. Figure 6The present invention provides a block diagram of the module structure of an intelligent claims processing device. The device includes: a data acquisition module 601, used to acquire multimodal claims application data in response to detecting a claims application from a target object; a disease association module 602, used to associate the multimodal claims application data with diseases based on a preset knowledge graph to obtain a target disease; wherein the preset knowledge graph is constructed based on the relationship between preset diseases, disease symptoms, and treatment plans; a health assessment module 603, used to perform a health assessment on the target object based on the multimodal claims application data and the target disease to obtain disease severity; a treatment assessment module 604, used to acquire the target symptoms of the target disease and perform a treatment assessment based on the multimodal claims application data and the target symptoms to obtain treatment rationality; a cost assessment module 605, used to acquire the target treatment plan for the target disease and perform a cost assessment based on the multimodal claims application data and the target treatment plan to obtain cost relevance; an insurance claims settlement module 606, used to fuse data based on disease severity, treatment rationality, and cost relevance to obtain claims settlement recommendations; and an insurance claims processing module 607, used to process claims based on the target disease and claims settlement recommendations.

[0082] It should be noted that the specific implementation method of this intelligent claims device is basically the same as the specific implementation method of the above-mentioned intelligent claims method, and will not be described again here.

[0083] This application also provides an electronic device, which includes: a memory, a processor, a program stored in the memory and executable on the processor, and a data bus for communication between the processor and the memory. When the program is executed by the processor, it implements the aforementioned intelligent claims method. This electronic device can be any intelligent terminal, including tablet computers, in-vehicle computers, etc.

[0084] Please see Figure 7 , Figure 7 The hardware structure of an electronic device according to another embodiment is illustrated. The electronic device includes: The processor 701 can be implemented using a general-purpose CPU (Central Processing Unit), microprocessor, application-specific integrated circuit (ASIC), or one or more integrated circuits, and is used to execute relevant programs to implement the technical solutions provided in the embodiments of this application. The memory 702 can be implemented as a read-only memory (ROM), a static storage device, a dynamic storage device, or a random access memory (RAM). The memory 702 can store the operating system and other applications. When the technical solutions provided in the embodiments of this specification are implemented through software or firmware, the relevant program code is stored in the memory 702 and is called and executed by the processor 701 using the intelligent claims method of the embodiments of this application. The input / output interface 703 is used to implement information input and output; The communication interface 704 is used to enable communication and interaction between this device and other devices. Communication can be achieved through wired means (such as USB, Ethernet cable, etc.) or wireless means (such as mobile network, WIFI, Bluetooth, etc.). Bus 705 transmits information between various components of the device (e.g., processor 701, memory 702, input / output interface 703, and communication interface 704); The processor 701, memory 702, input / output interface 703, and communication interface 704 are connected to each other within the device via bus 705.

[0085] This application also provides a computer-readable storage medium for computer-readable storage, wherein the storage medium stores one or more programs, which can be executed by one or more processors to implement the above-described intelligent claims method.

[0086] Memory, as a non-transitory computer-readable storage medium, can be used to store non-transitory software programs and non-transitory computer-executable programs. Furthermore, memory may include high-speed random access memory, and may also include non-transitory memory, such as at least one disk storage device, flash memory device, or other non-transitory solid-state storage device. In some embodiments, memory may optionally include memory remotely located relative to the processor, and these remote memories can be connected to the processor via a network. Examples of such networks include, but are not limited to, the Internet, intranets, local area networks, mobile communication networks, and combinations thereof.

[0087] The embodiments described in this application are for the purpose of more clearly illustrating the technical solutions of the embodiments of this application, and do not constitute a limitation on the technical solutions provided by the embodiments of this application. As those skilled in the art will know, with the evolution of technology and the emergence of new application scenarios, the technical solutions provided by the embodiments of this application are also applicable to similar technical problems.

[0088] Those skilled in the art will understand that the technical solutions shown in the figures do not constitute a limitation on the embodiments of this application, and may include more or fewer steps than shown, or combine certain steps, or different steps.

[0089] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0090] Those skilled in the art will understand that all or some of the steps in the methods disclosed above, as well as the functional modules / units in the systems and devices, can be implemented as software, firmware, hardware, or suitable combinations thereof.

[0091] The terms “first,” “second,” “third,” “fourth,” etc. (if present) in the specification and accompanying drawings of this application 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 this application 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.

[0092] It should be understood that in this application, "at least one (item)" means one or more, and "more than" means two or more. "And / or" is used to describe the relationship between related objects, indicating that three relationships can exist. For example, "A and / or B" can represent three cases: only A exists, only B exists, and both A and B exist simultaneously, where A and B can be singular or plural. The character " / " generally indicates that the preceding and following related objects are in an "or" relationship. "At least one (item) of the following" or similar expressions refer to any combination of these items, including any combination of single or plural items. For example, at least one (item) of a, b, or c can represent: a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, and c can be single or multiple.

[0093] In the several embodiments provided in this application, it should be understood that the disclosed apparatus and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or units may be electrical, mechanical, or other forms.

[0094] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0095] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0096] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes multiple instructions to cause an electronic device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing programs, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0097] It should be noted that any AI models, software tools, or components not belonging to this company appearing in the embodiments of this application are merely illustrative examples and do not represent actual use. All user personal information involved in the embodiments of this application has been authorized (with the knowledge and consent) by the relevant parties or has been fully authorized by all parties, and the executing entity may obtain it through various legal and compliant means. The collection, storage, use, processing, transmission, provision, and disclosure of the information, data, and signals involved all comply with relevant laws and regulations and do not violate public order and good morals.

[0098] The preferred embodiments of the present application have been described above with reference to the accompanying drawings, but this does not limit the scope of the claims of the present application. Any modifications, equivalent substitutions, and improvements made by those skilled in the art without departing from the scope and substance of the embodiments of the present application shall be within the scope of the claims of the present application.

Claims

1. An intelligent claims settlement method, characterized in that, The method includes: In response to the detection of a claim application from a target object, multimodal claim application data is obtained; Disease association is performed on the multimodal claims data based on a preset knowledge graph to obtain the target disease; wherein, the preset knowledge graph is constructed based on the relationship between preset diseases, disease symptoms, and treatment plans; Based on the multimodal claims data and the target disease, a health assessment is performed on the target individual to obtain the disease severity. Obtain the target symptoms of the target disease, and conduct a treatment assessment based on the multimodal claim data and the target symptoms to determine the rationality of the treatment; Obtain the target treatment plan for the target disease, and conduct a cost assessment based on the multimodal claim data and the target treatment plan to obtain the cost correlation. Based on the severity of the disease, the reasonableness of the treatment, and the relevance of the costs, a claims assessment recommendation is obtained. The claim will be processed based on the target disease and the claim assessment recommendation.

2. The method according to claim 1, characterized in that, The method is applied to a claims system, which includes a managed claims platform, a digital intelligence platform, and a data middleware. The step of obtaining multimodal claim data in response to detecting a claim application from a target object includes: In response to the detection of a claim application from a target object, the insurance case of the target object is determined through the managed claims platform, and the case identifier of the insurance case is transmitted to the digital platform; The system receives the case identifier through the digital intelligence platform and sends a case data request to the data platform based on the case identifier. Through the data platform, data is read from at least two data sources based on the case data request to obtain multimodal claim application data, and the multimodal claim application data is transmitted to the digital intelligence platform.

3. The method according to claim 2, characterized in that, The claims system also includes an intelligent diagnostic engine, which is bound to the preset knowledge graph; The process of associating the multimodal claims data with diseases based on a preset knowledge graph to obtain target diseases includes: The digital intelligence platform generates a disease reasoning task based on the multimodal claims application data and transmits the disease reasoning task to the intelligent diagnostic engine. The intelligent diagnostic engine matches the target disease with the preset knowledge graph and the multimodal claims application data.

4. The method according to claim 3, characterized in that, The step of matching the target disease based on the preset knowledge graph and the multimodal claims application data includes: The multimodal claims data is fused using a large language model to obtain disease description text; Based on the disease description text, the disease symptoms and treatment plans of candidate diseases in the preset knowledge graph are matched to obtain the matching confidence of the candidate diseases; The candidate diseases are screened based on the matching confidence level to obtain the target disease.

5. The method according to claim 3, characterized in that, Before performing disease association on the multimodal claims application data based on a preset knowledge graph to obtain the target disease, the method further includes: Obtain raw medical data, including raw disease data, raw symptom data, and raw treatment plan data; Entity extraction is performed based on the original disease data to obtain disease nodes, entity extraction is performed based on the original symptom data to obtain symptom nodes, and entity extraction is performed based on the original treatment plan data to obtain treatment plan nodes. Create a first edge between the disease node and the symptom node, create a second edge between the disease node and the treatment plan, and create a third edge between the symptom node and the treatment plan; Using medical literature evidence, attention is calculated between the disease node and the symptom node to obtain the first association weight of the first side; attention is calculated between the disease node and the treatment plan node to obtain the second association weight of the second side; and attention is calculated between the symptom node and the treatment plan node to obtain the third association weight of the third side, so as to construct the preset knowledge graph. The preset knowledge graph is bound to the intelligent diagnostic engine.

6. The method according to claim 5, characterized in that, The claims system also includes a multi-source monitoring agent platform. After binding the preset knowledge graph to the intelligent diagnostic engine, the method further includes: The multi-source monitoring agent platform detects the medical database, and when an update is detected in the medical database, it acquires the new medical data and transmits the new medical data to the intelligent diagnostic engine. The intelligent diagnostic engine updates the preset knowledge graph based on the newly added medical data.

7. The method according to claim 2, characterized in that, The claims system also includes a claims review rule engine, which performs a health assessment on the target individual based on the multimodal claims application data and the target disease to obtain the disease severity, including: The claims review rule engine extracts pathology reports and imaging reports from the multimodal claims application data. The claims processing rule engine is used to assess disease progression based on the pathology report, the imaging report, and the target disease, thereby obtaining the disease progression level of the target individual. The treatment tolerance score of the target subject is obtained by using the claims review rule engine based on the pathology report, the imaging report and the target disease to conduct a treatment assessment. The disease severity is obtained by fusing the disease progression level and the treatment tolerance score.

8. An intelligent claims processing device, characterized in that, The device includes: The data acquisition module is used to acquire multimodal claim application data in response to the detection of a claim application from a target object; The disease association module is used to associate diseases in the multimodal claims application data based on a preset knowledge graph to obtain the target disease; wherein, the preset knowledge graph is constructed based on the relationship between preset diseases, disease symptoms, and treatment plans; The health assessment module is used to perform a health assessment on the target object based on the multimodal claim application data and the target disease, and to obtain the severity of the disease; The treatment assessment module is used to obtain the target symptoms of the target disease, and to conduct a treatment assessment based on the multimodal claim data and the target symptoms to determine the rationality of the treatment. The cost assessment module is used to obtain the target treatment plan for the target disease, and to assess the cost based on the multimodal claim data and the target treatment plan to obtain the cost correlation. The insurance claims module is used to fuse data based on the severity of the disease, the reasonableness of the treatment, and the relevance of the cost to obtain claims recommendations. The insurance claims module is used to process the claims application based on the target disease and the claims review recommendation.

9. An electronic device, characterized in that, The electronic device includes a memory and a processor, the memory storing a computer program, and the processor executing the computer program to implement the method according to any one of claims 1 to 7.

10. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by a processor, it implements the method of any one of claims 1 to 7.

Images