CN-121304354-B - Intelligent business supervision and behavior tracing system and method based on data driving

CN121304354BCN 121304354 BCN121304354 BCN 121304354BCN-121304354-B

Abstract

The invention relates to the technical field of data processing, and discloses a business intelligent supervision and behavior tracing system and method based on data driving, wherein the method comprises the steps of obtaining basic data of medical services of patients to construct a multidimensional association map; the method comprises the steps of extracting a diagnosis path pool of a disease generated by a patient group with the same diagnosis, constructing a point set of a diagnosis track coordinate system, generating a golden diagnosis path template, matching and aligning a real-time diagnosis path of a target patient with the template, judging whether the target patient is an abnormal path, and if so, carrying out traceable analysis.

Inventors

YAN LIEHU

Assignees

浙江红阵易审数据科技有限公司

Dates

Publication Date: 20260505
Application Date: 20251016

Claims (6)

1. The intelligent business supervision and behavior tracing method based on data driving is characterized by comprising the following steps: The method comprises the steps of obtaining basic data of patient medical service, wherein the basic data of patient medical service at least comprises patient diagnosis code information, diagnosis and treatment item time sequence records and medical insurance settlement information, constructing a preliminary diagnosis association diagram based on the patient diagnosis code information, extracting diagnosis codes from the patient diagnosis code information as nodes, adding diagnosis and treatment item nodes established according to the diagnosis and treatment item time sequence records and medical insurance settlement nodes established according to the medical insurance settlement information around the core nodes in the constructed preliminary diagnosis association diagram by taking the diagnosis codes as the core nodes, wherein the diagnosis code core nodes are respectively connected with the diagnosis and treatment item nodes and the medical insurance settlement nodes, and the medical insurance settlement nodes are connected with the corresponding diagnosis and treatment item nodes to construct a multidimensional association diagram for comprehensively showing diagnosis and treatment processes; Extracting a co-diagnosis patient group from the multidimensional correlation map, and extracting diagnosis and treatment project time sequence records of each patient based on the co-diagnosis patient group to generate a disease diagnosis and treatment path pool; Obtaining a first diagnosis and treatment time point from the diagnosis and treatment project time sequence record, constructing a diagnosis and treatment track coordinate system by taking the first diagnosis and treatment time point as an origin O, and mapping the standardized diagnosis and treatment track vector V' into the diagnosis and treatment track coordinate system to obtain a diagnosis and treatment track point set; Calculating the local density of each point in the diagnosis and treatment track point set, identifying the point with the highest density, calculating the distance between each point in the diagnosis and treatment track point set and the point with the highest density, identifying the core track point according to the local density and the distance, connecting the core track points in a diagnosis and treatment track coordinate system to form a main path, and incorporating the diagnosis and treatment track points with the local density around each core track point being greater than a preset density threshold value into corresponding branch paths to form a gold diagnosis and treatment path template; acquiring a real-time diagnosis and treatment path of a target patient, matching and matching the real-time diagnosis and treatment path with a gold diagnosis and treatment path template, and judging whether the real-time diagnosis and treatment path of the target patient is an abnormal diagnosis and treatment path or not; and if the abnormal diagnosis and treatment path is judged, performing traceability analysis on the abnormal diagnosis and treatment path.
2. The method for intelligently supervising and tracing business based on data driving according to claim 1, wherein the method for constructing a preliminary diagnosis association graph based on patient diagnosis code information comprises adding a directed edge to connect two diagnosis code nodes if two diagnosis codes occur simultaneously in more than Y patients and there is a sequence, wherein Y is a preset threshold parameter.
3. The data-driven business intelligent supervision and behavior trace-back method according to claim 2, wherein the method for extracting the co-diagnostic patient group from the multidimensional correlation map comprises the following steps: Determining a target diagnostic code, setting a diagnostic similarity radius r by taking the target diagnostic code as a center in a multidimensional correlation map, and constructing a diagnostic neighbor set, wherein the diagnostic neighbor set comprises all diagnostic codes which are not more than r from the target diagnostic code; The method comprises the steps of collecting diagnostic records corresponding to all diagnostic codes in a diagnostic neighbor set to form an original diagnostic group, screening complications and complications from the original diagnostic group based on a multidimensional correlation map, and obtaining a pure diagnostic group; calculating the diagnostic purity index of the pure diagnostic group; If the diagnostic purity index is greater than the preset purity threshold, the purity diagnostic group is identified as the same diagnostic patient group.
4. The method for intelligently supervising and tracing business based on data driving according to claim 3, wherein the method for calculating the diagnostic purity index of the pure diagnostic group comprises: Counting the number of diagnostic records corresponding to all diagnostic codes in the diagnostic neighbor set, marking the diagnostic records as the total number of diagnostic records, counting the number of screened complications and complications diagnostic records, and dividing the diagnostic purity index by the total number of diagnostic records after subtracting the number of screened complications and complications diagnostic records from the total number of diagnostic records.
5. The data-driven business intelligent supervision and behavior tracing method according to claim 4, wherein the diagnosis and treatment project time sequence record at least comprises diagnosis and treatment project codes, diagnosis and treatment project cost and specific implementation time; The method for constructing the standardized diagnosis and treatment track vector V' according to the diagnosis and treatment item time sequence records of patients in the disease diagnosis and treatment path pool comprises the following steps: Taking the first diagnosis and treatment time point as a datum point t 0 , calculating a time interval sequence delta t of the specific implementation time of each follow-up diagnosis and treatment project relative to t 0 ; mapping the diagnosis and treatment item codes into standardized diagnosis and treatment item codes, and generating a standardized diagnosis and treatment item sequence P; Calculating the ratio of the cost of each diagnosis and treatment project to the average cost of the same project in the same region at the same time, and generating a cost standardization coefficient sequence C; Combining the time interval sequence deltat, the standardized diagnosis and treatment item sequence P and the cost standardization coefficient sequence C into a three-dimensional diagnosis and treatment track vector V= [ deltat, P and C ], and executing dimension normalization processing on the three-dimensional diagnosis and treatment track vector V to obtain a standardized diagnosis and treatment track vector V'.
6. A data-driven-based business intelligent supervision and behavior tracing system for implementing the data-driven-based business intelligent supervision and behavior tracing method according to any one of claims 1 to 5, characterized in that the system comprises: the map construction module is used for acquiring basic data of medical services of patients and constructing a multidimensional association map for comprehensively showing the diagnosis and treatment process; The disease diagnosis and treatment path pool generation module is used for extracting the patient group with the same diagnosis from the multidimensional association map to generate a disease diagnosis and treatment path pool; The diagnosis and treatment track generation module is used for constructing a diagnosis and treatment track coordinate system according to the disease diagnosis and treatment path pool to obtain a diagnosis and treatment track point set; The path template generation module is used for generating a gold diagnosis and treatment path template based on the diagnosis and treatment track point set; The abnormal path judging module is used for acquiring a real-time diagnosis and treatment path of the target patient, matching and aligning the real-time diagnosis and treatment path with the gold diagnosis and treatment path template, and judging whether the real-time diagnosis and treatment path of the target patient is an abnormal diagnosis and treatment path or not; the tracing analysis module is used for carrying out tracing analysis on the abnormal diagnosis and treatment path.

Description

Intelligent business supervision and behavior tracing system and method based on data driving Technical Field The invention relates to the technical field of data processing, in particular to a business intelligent supervision and behavior tracing system and method based on data driving. Background The medical insurance business supervision has important significance in maintaining medical business order and guaranteeing the safety of the foundation. Along with the increase of medical data volume and the complicating of diagnosis and treatment behaviors, the traditional supervision mode faces a plurality of challenges, and an intelligent means is needed to improve the supervision efficiency, so that the fine management of medical insurance business is realized. The reasonable use of the medical insurance foundation directly relates to the sustainable development of medical insurance business and the long-term stability of a medical insurance system, and the intelligent degree of a medical insurance supervision system has become a key index for measuring the modernization level of the medical insurance business. The Chinese patent with publication number of CN119048189A provides an intelligent supervision system, method and electronic equipment based on medical insurance data, which provide supervision support for medical insurance purchasing of a pharmacy through a data storage and screening mechanism. The system mainly focuses on compliance examination of medicine proportion, but lacks deep analysis capability of diagnosis and treatment full-flow service modes, and cannot combine medical insurance supervision with intelligent service decision, so that prospective guidance cannot be provided for medical insurance service from a data driving angle. The Chinese patent with publication number CN119579332A relates to medical insurance general outpatient service supervision system and method based on fraud abnormal portrait technology, and uses modules of medical transaction data acquisition, behavior analysis, risk assessment and the like to realize accurate identification and prevention of fraud. The system focuses on the detection of abnormal behaviors, but the analysis method depends on a preset rule base, so that the system is difficult to adapt to the dynamic change of a medical insurance service mode, lacks the intelligent analysis capability of a data-driven service, and cannot realize the real-time intervention of the medical insurance service. However, in actual medical insurance business operation, the patient with the same diagnosis disease may present diversified diagnosis and treatment paths due to individual differences, disease complexity and different judgment of doctors, wherein part of paths may deviate from a reasonable range, and illegal actions such as excessive examination, repeated treatment or treatment avoidance are involved, so that medical insurance funds are wasted. The traditional rule base is difficult to dynamically identify abnormal paths hidden in reasonable diagnosis and treatment differences due to static preset characteristics. For example, in a diagnosis and treatment path, part of unnecessary examination may be disguised as a routine examination item, and due to lack of a real-time dynamic comparison and analysis means, a supervision system cannot find such abnormalities in time, so that a medical insurance fund is lost, and sustainable development of medical insurance business is affected. The medical insurance business supervision needs to be changed from "post-responsibility-pursuing" to "in-process intervention" and updated from "static rules" to "dynamic intelligence", so as to realize intelligent management of the medical insurance business process. Disclosure of Invention In order to overcome the defects of the prior art, the invention provides the business intelligent supervision and behavior tracing system and method based on data driving, which are used for accurately identifying the abnormal diagnosis and treatment behaviors by constructing a multidimensional association map, generating a gold diagnosis and treatment path template and comparing the patient diagnosis and treatment paths in real time, so that the problem that the abnormal treatment paths under the condition of 'same disease and different treatments' are difficult to find in real time in the prior art is effectively solved, the reasonable use of medical insurance funds is ensured, and the fairness and the sustainability of medical insurance degree are maintained. In order to achieve the above purpose, the present invention provides the following technical solutions: A business intelligent supervision and behavior tracing method based on data driving comprises the following steps: acquiring basic data of medical services of patients, and constructing a multidimensional association map for comprehensively showing diagnosis and treatment processes; Extracting the patient group to be diagnosed from the