WO2020119403A1

WO2020119403A1 - Hospitalization data abnormity detection method, apparatus and device, and readable storage medium

Info

Publication number: WO2020119403A1
Application number: PCT/CN2019/119215
Authority: WO
Inventors: 陈明东; 黄越; 胥畅
Original assignee: 平安医疗健康管理股份有限公司
Priority date: 2018-12-13
Filing date: 2019-11-18
Publication date: 2020-06-18
Also published as: CN109635044A

Abstract

The present application relates to the technical field of big data, and provides a hospitalization data abnormity detection method, comprising: obtaining clinic hospitalization data of insured personnel; inputting the clinic hospitalization data into a preset model, and determining whether abnormal data of short hospitalization examination exists in the clinic hospitalization data or not based on the preset model; and if yes, determining a fee deduction basis based on the inspection details of the short hospitalization examination in the abnormal data, so as to send a fee deduction notification to a corresponding designated medical institution based on the fee deduction basis. Also disclosed are a hospitalization data abnormity detection apparatus and device and a readable storage medium. The present application realizes management of outpatient overall fund expenditure for each human resources and social security bureau, and prevents unreasonable outpatient expenditure.

Description

Hospitalization data abnormality detection method, device, equipment and readable storage medium The

This application requires the priority of the Chinese patent application filed on December 13, 2018 in the Chinese Patent Office with the application number 201811530944.4 and the invention titled "Hospitalization Data Anomaly Detection Method, Device, Equipment, and Readable Storage Media". Incorporated by reference in the application.

Technical field

The present application relates to the field of big data technology, and in particular to a method, device, equipment, and readable storage medium for hospitalization data anomaly detection.

Background technique

Medical insurance generally refers to basic medical insurance, a social insurance system established to compensate workers for economic losses caused by disease risks. The medical insurance fund shall be established by the employer and the individual to pay the fees. After the insured person incurs medical expenses, the medical insurance institution shall give him certain economic compensation. In the prior art, supervisors are equipped to monitor the social insurance behavior of the insured personnel, and to calculate the medical insurance settlement documents in order to manage the outpatient co-ordination fund expenditure. However, due to the large number of medical insurance insurers, relying solely on supervisors to supervise the insured data is not enough, for example, there may be short hospitalizations in the insured medical data for inspections, which will greatly cause outpatient co-ordination funds. waste.

Summary of the invention

The main purpose of this application is to provide a method, device, equipment and readable storage medium for anomaly detection of hospitalization data, which aims to solve the problem that the existing technology only relies on supervisors to supervise the insured data, and the strength is not enough, resulting in outpatient co-ordination fund management Defective technical problems.

To achieve the above purpose, the present application provides a method for detecting abnormality of hospitalization data. The method for detecting abnormality of hospitalization data includes:

Obtain medical data of insured persons;

Determine whether there is abnormal data for short hospitalization for examination in the hospitalization data based on a preset model;

If yes, the basis for the deduction is determined based on the details of the short hospitalization in the abnormal data, so that a deduction notice is sent to the corresponding designated medical institution based on the deduction basis.

Optionally, the step of obtaining inpatient data of insured persons includes:

Obtain the hospitalization data of the insured personnel uploaded by the designated medical institution, where the hospitalization data at least includes the designated medical institution, information of the insured person, the reason for hospitalization, the number of days admitted, surgery follow-up, transfer records and hospitalization examination costs .

Optionally, the step of determining whether there is abnormal data for short hospitalization for examination based on a preset model includes:

Convert the text data in the hospitalization data into standard data;

Perform cluster analysis on the standard data based on a clustering algorithm, and determine whether abnormal data exists in the standard data for short hospitalization based on the result of the cluster analysis.

Optionally, the step of converting the text data in the hospitalization data into standard data includes:

Clean the hospitalization data to obtain the standardized data after cleaning;

Construct text features based on the specification data to obtain corresponding word vectors;

The word vector is encoded into a vector matrix based on a bidirectional recurrent neural network RNN model, so as to determine corresponding standard data based on the vector matrix.

Optionally, after the step of converting the text data in the hospitalization data into standard data, the method further includes:

Determine whether the standard data needs to be reduced in dimension;

If yes, perform dimensionality reduction processing on the standard data based on the linear method, and perform cluster analysis on the standard data based on the clustering algorithm, and determine whether the standard data is based on the result of the cluster analysis There is a procedure of abnormal data for short hospitalization for examination.

Optionally, the step of performing cluster analysis on the standard data based on a clustering algorithm and judging whether there is abnormal data in the standard data for short hospitalization based on the result of the cluster analysis includes:

Perform cluster analysis on the standard data based on a clustering algorithm, and divide the standard data into different categories, where the standard data includes at least the number of days of admission and the cost of hospitalization examination;

Judging whether there is standard data for the number of admission days less than the preset threshold in the different categories;

If yes, it is judged whether the inpatient examination cost corresponding to the standard data of the number of admission days is less than the preset number of days threshold is greater than the preset cost threshold.

Optionally, the step of determining the basis for deduction based on the details of the short hospitalization in the abnormal data so that the step of sending a deduction notice to the corresponding designated medical institution based on the basis for deduction includes:

Obtain a preset fee deduction rule, and based on the preset fee deduction rule, and the difference in the number of days between the admission days and the preset number of days threshold in the abnormal data, and the The cost difference determines the basis for deduction;

Generate a deduction notice based on the deduction basis, and send the deduction notice to the corresponding designated medical institution.

In addition, in order to achieve the above object, the present application also provides a hospitalization data abnormality detection device, which includes:

The hospitalization data acquisition module is used to obtain the hospitalization data of the insured personnel;

The abnormal data judgment module is used to judge whether there is abnormal data for short hospitalization for examination based on a preset model;

The deduction basis determination module is used to determine the deduction basis based on the inspection details of the short hospitalization in the abnormal data if it is so as to send a deduction notice to the corresponding designated medical institution based on the deduction basis.

In addition, in order to achieve the above object, the present application also provides a hospitalization data abnormality detection device, the hospitalization data abnormality detection device includes: a memory, a processor, and a computer stored on the memory and operable on the processor Readable instructions, which when executed by the processor, implement the steps of the hospitalization data anomaly detection method described above.

In addition, in order to achieve the above object, the present application also provides a readable storage medium having computer readable instructions stored thereon, the computer readable instructions being executed by the processor to realize the hospitalization data as described above Anomaly detection method steps.

An abnormality detection method for hospitalization data proposed in this application first obtains the hospitalization data of insured persons, and judges whether there is abnormal data of short hospitalization for examination in the hospitalization data based on a preset model, if it exists, it is based on the abnormal data Check the details of the short-term hospitalization to determine the basis for deduction, so as to send the deduction notice to the corresponding designated medical institution based on the basis for deduction. Through the hospitalization data anomaly detection method proposed in this application, the insured personnel's hospitalization inspection is detected to determine whether there is abnormal data for short hospitalization examinations, and deductions are sent to designated medical institutions that have short hospitalization examination abnormality data Notice to prevent unreasonable outpatient expenses, effectively manage the outpatient co-ordination fund expenditure, and avoid wasting the outpatient co-ordination fund.

BRIEF DESCRIPTION

FIG. 1 is a schematic diagram of the hardware structure of a hospitalization data abnormality detection device involved in an embodiment of the present application;

FIG. 2 is a schematic flowchart of a first embodiment of a method for detecting abnormality in hospital application data;

FIG. 3 is a detailed flowchart of step S20 in FIG. 2;

FIG. 4 is a schematic flowchart of a second embodiment of a method for detecting abnormality in hospitalization data of the application;

FIG. 5 is a schematic diagram of functional modules of a first embodiment of an apparatus for detecting abnormality of hospitalization data of the application.

The implementation, functional characteristics and advantages of the present application will be further described in conjunction with the embodiments and with reference to the drawings.

detailed description

It should be understood that the specific embodiments described herein are only used to explain the present application, and are not used to limit the present application.

The main solution of the embodiment of the present application is to obtain the inpatient hospitalization data of the insured personnel; determine whether there is abnormal data of short hospitalization for examination in the hospitalization data based on a preset model; if it is, based on the abnormal data in the abnormal data Check the details of the short hospital stay to determine the basis for the fee deduction, so as to send the fee deduction notice to the corresponding designated medical institution based on the basis for the fee deduction. The technical solution of the embodiment of the present application solves the technical problem in the prior art that only relying on supervisors to supervise the insured data, the strength is not enough, and the outpatient co-ordination fund management has defects.

As shown in FIG. 1, FIG. 1 is a schematic diagram of a hardware structure of a hospitalization data abnormality detection device involved in an embodiment of the present application.

The hospitalization data abnormality detection method according to the embodiments of the present application is mainly applied to hospitalization data abnormality detection equipment. The hospitalization data abnormality detection equipment may be a device with display and processing functions such as a PC, a portable computer, and a mobile terminal.

As shown in FIG. 1, the hospitalization data abnormality detection device may include: a processor 1001, such as a CPU, a communication bus 1002, a user interface 1003, a network interface 1004, and a memory 1005. Among them, the communication bus 1002 is used to implement connection communication between these components. The user interface 1003 may include a display (Display), an input unit such as a keyboard (Keyboard), and the optional user interface 1003 may also include a standard wired interface and a wireless interface. The network interface 1004 may optionally include a standard wired interface and a wireless interface (such as a WI-FI interface). The memory 1005 may be a high-speed RAM memory or a stable memory (non-volatile memory), such as disk storage. The memory 1005 may optionally be a storage device independent of the foregoing processor 1001.

Optionally, the hospitalization data abnormality detection device may also include a camera, RF (Radio Frequency (radio frequency) circuits, sensors, audio circuits, Wi-Fi modules, etc. Among them, sensors such as light sensors, motion sensors and other sensors. Of course, the hospitalization data abnormality detection device can also be equipped with other sensors such as gyroscopes, barometers, hygrometers, thermometers, infrared sensors, and so on, which will not be repeated here.

Those skilled in the art may understand that the structure of the hospitalization data abnormality detection device shown in FIG. 1 does not constitute a limitation on the hospitalization data abnormality detection device, and may include more or fewer components than the illustration, or a combination of certain components, Or different component arrangements.

As shown in FIG. 1, the memory 1005 as a computer storage medium may include an operating system, a network communication module, a user interface module, and computer-readable instructions. In FIG. 1, the network interface 1004 is mainly used to connect to a background server and perform data communication with the background server; the user interface 1003 is mainly used to connect to a client (user side) and perform data communication with the client; and the processor 1001 and the memory 1005 It may be provided in a hospitalization data abnormality detection device that calls the computer-readable instructions stored in the memory 1005 through the processor 1001 and executes the hospitalization data abnormality detection method provided by the embodiment of the present application.

The solution provided in this embodiment first obtains the inpatient hospitalization data of the insured person, and judges whether there is abnormal data of the short hospitalization in the hospitalization data based on the preset model, and if it exists, performs the inspection based on the short hospitalization in the abnormal data Determine the basis for deduction in detail so that the deduction notice will be sent to the corresponding designated medical institution based on the basis for deduction. Through the hospitalization data abnormality detection method proposed in this application, the insured personnel's hospitalization inspection is detected to determine whether there is abnormal data for short hospitalization examinations, and deductions are issued for designated medical institutions that have short hospitalization examination abnormality data Notice to prevent unreasonable outpatient expenses, effectively manage the outpatient co-ordination fund expenditure, and avoid wasting the outpatient co-ordination fund.

Based on the above hardware structure, an embodiment of the method for detecting abnormality of hospitalization data of the present application is proposed.

Referring to FIG. 2, FIG. 2 is a schematic flowchart of a first embodiment of a method for detecting abnormality of hospitalization data of the present application. In this embodiment, the method includes:

Step S10: Obtain medical data of the insured personnel;

In the prior art, when the insured personnel use the medical insurance card to settle the medical expenses in designated medical institutions, such as hospitals and pharmacies, the charging terminal equipment of the designated medical institutions can upload the corresponding medical data to the core system of the human society. The core system of the People's Society checks the corresponding medical data of the insured personnel to manage the outpatient co-ordination fund expenditure. In this embodiment, the in-hospital inspection of the insured personnel is detected. Therefore, the medical consultation data in this embodiment specifically refers to the medical admission hospitalization data.

Specifically, the hospitalization data includes at least the designated medical institution, information of the insured person, days of admission, follow-up of surgery, transfer records, and hospitalization expenses. Among them, the information of the insured person can specifically include the name of the insured person, Age and medical insurance card information of the insured.

Step S20, based on a preset model, it is determined whether there is abnormal data in the hospitalization data for short hospitalization for examination; if so, step S30 is executed;

In this embodiment, it is judged whether the inpatient data of the insured person has abnormal data of short hospitalization for examination by putting the inpatient data in the preset model to run. Specifically, the data used in this embodiment The preset model is mainly a deviation detection model. The deviation specifically refers to anomalous instances in the classified samples, special cases that do not meet the rules, or the observation results are inconsistent with the model predictions and the observation results change with time. The basic goal of deviation detection is to find There is a meaningful difference between the observation and the reference value.

Specifically, as shown in FIG. 3, the step S20 specifically includes:

Step S21, converting the text data in the hospitalization data into standard data;

There is some unnecessary information in the hospitalization data uploaded to the core system of the human society, which is not conducive to the model's processing of the data. Therefore, first, the original hospitalization data is processed through the NLP natural language processing process and converted into Standard data.

Specifically, the hospitalization data is input into the deviation detection model, and the hospitalization data is cleaned. There are some unnecessary data or non-standard data in the hospitalization data. The data cleaning is mainly to delete the irrelevant data, duplicate data in the original data set, smooth the noise data, and filter out the data that is not related to the subject of the model detection, and deal with the missing values. , Outliers, etc. For example, the information included in the hospitalization data that is not related to the hospitalization examination, such as auxiliary medication information. Through data cleaning, standard data corresponding to the original hospitalization data is obtained.

Further, text features are constructed based on the obtained canonical data to obtain corresponding word vectors. First, use RNN (Recurrent Neural Network (recurrent neural network) model performs word segmentation processing and removal of useless words on the cleaned hospitalization data. For Chinese text data, such as a sentence containing Chinese, the words are continuous between words, and the minimum unit granularity of data analysis is words. Therefore, it is necessary to perform word segmentation on the hospitalization data. Further, the useless words are removed from the data after word segmentation. Useless words are words that do not contribute anything to the features of the text, such as: ah, yes, yes, you, me, and of course, some punctuation marks, These useless words do not reflect the meaning of the text, therefore, need to remove the useless words. After word segmentation processing and removal of useless words, the text data corresponding to the hospitalization data can be obtained, so that the text features can be constructed according to the text data to further obtain the corresponding word vector.

Construct text features based on the text data corresponding to the treated hospitalization data, represent the text as a sequence of vectors, and express the words as word vectors, and obtain the complete word vector by stitching forward and reverse calculations to further determine the word vector Sequence, and then use the bidirectional RNN model to encode the word vector into a sentence vector matrix in order to get the final standard data from the sentence vector matrix. In this embodiment, the standard data may include the number of days of admission, the reason for hospitalization, the cost of hospitalization, follow-up surgery, and transfer records. For example, if an insured person is admitted to the hospital for treatment due to pneumonia, the standard data of his hospitalization may include the number of days of hospitalization, pneumonia due to hospitalization, and a few yuan for hospitalization.

Step S22, performing cluster analysis on the standard data based on a clustering algorithm, and judging whether there is abnormal data for short hospitalization examination in the standard data based on the result of the cluster analysis.

Perform cluster analysis on the above obtained criteria according to the clustering algorithm. Specifically, the k-means algorithm can be used to implement data clustering, so that the obtained cluster meets: the similarity of objects in the same cluster is high, and The similarity of objects in different clusters is small, and the clustering results are visualized to determine whether there is abnormal data for short hospitalization for examination.

Step S30: Determine the basis of the fee deduction based on the details of the short hospitalization in the abnormal data, so as to send a fee deduction notice to the corresponding designated medical institution based on the basis of the deduction.

In this embodiment, if abnormal data for short hospitalization is detected in the hospitalization data, the preset deduction rule is obtained, and the basis for deduction is determined according to the predetermined deduction rule and the abnormal data. It can be understood that For different abnormal data, there can be different preset deduction rules.

Further, a deduction notice is generated based on the deduction basis, and the deduction notice may include the designated medical institution for deduction penalties, the reason for the deduction, the deduction measures and the time, etc., and send the deduction notice to the corresponding fixed point Medical institutions in order to punish them.

In this embodiment, first obtain the hospitalization data of the insured personnel, and determine whether there is abnormal data of the short hospitalization for examination in the hospitalization data based on the preset model, and if so, based on the details of the short hospitalization in the abnormal data Determine the basis for the deduction so that the deduction notice will be sent to the corresponding designated medical institution based on the basis for the deduction. Through the hospitalization data abnormality detection method proposed in this application, the insured personnel's hospitalization inspection is detected to determine whether there is abnormal data for short hospitalization examinations, and deductions are issued for designated medical institutions that have short hospitalization examination abnormality data Notice to prevent unreasonable outpatient expenses, effectively manage the outpatient co-ordination fund expenditure, and avoid wasting the outpatient co-ordination fund.

Further, referring to FIG. 4, based on the foregoing embodiment, a second embodiment of the method for detecting abnormality of hospitalization data of the present application is proposed. In this embodiment, after step S21, the method further includes:

Step S23, judging whether the standard data needs dimensionality reduction processing; if so, step S24 is executed;

The dimensionality reduction processing of the standard data is to further extract and synthesize effective information and discard the useless information. First, judge whether the standard data needs dimensionality reduction processing, specifically, convert the text features corresponding to the standard data into vectors When expressing the numerical features in matrix form, the dimension corresponding to the vector matrix is the dimension corresponding to the standard data of hospitalization. If the dimension is too high, dimensionality reduction processing is required.

Step S24: Perform dimensionality reduction processing on the standard data based on a linear method, and execute step S22.

In this embodiment, the linear method is mainly used to reduce the dimensionality of multi-dimensional data. In the prior art, the commonly used dimensionality reduction methods of multi-dimensional data processing mainly include the following: missing value ratio, low variance filtering, and high correlation Filtering, random forest/combination tree, principal component analysis, reverse feature removal and forward feature construction. By performing dimensionality reduction processing on multidimensional data, it is possible to achieve dimensionality reduction on standard data for hospitalization.

Further, the standard data after dimensionality reduction processing is clustered according to the clustering algorithm. Specifically, the k-means algorithm can be used to implement data clustering as follows:

The k-means algorithm accepts the input amount k, and then divides n data objects into k clusters, so that the obtained clusters satisfy: objects in the same cluster have high similarity, but objects in different clusters are similar Degree is smaller. In this embodiment, the input amount K refers to the number of hospital admission days, the cost of inpatient examination, follow-up surgery, and transfer records in the standard data of hospitalization, and n data objects refer to all standard data to be classified. The specific process is as follows: First, k objects are randomly selected from the n data objects as the initial clustering centers, and for the remaining other objects, according to their similarity with these initial clustering centers, they are assigned to The cluster represented by the most similar cluster center, and then calculate the cluster center of each new cluster obtained, that is, the mean of all objects in the cluster, and repeat this process until the standard measurement function begins to converge, Generally, mean square error is used as the standard measurement function. The k clusters have the following characteristics: each cluster itself is as compact as possible, and each cluster is separated as much as possible.

Through the above clustering algorithm, the standard data is divided into several different categories based on the number of hospital admission days, hospitalization examination costs, follow-up surgery, transfer records, etc. And visualize the clustering results, in order to determine whether there is abnormal data in the hospitalization data for short hospitalization for examination.

Divide the standard data into different classes, and determine whether there are classes containing abnormal data in the different classes. Specifically, the category containing abnormal data refers to data in which the number of admission days in a certain cluster is less than the preset number of days threshold, but the cost of inpatient examination is greater than the conventional preset cost threshold. For example, the insured person who was admitted to the hospital due to pneumonia as mentioned above, if the standard number of hospital admissions is less than the preset number of days and the hospitalization cost is greater than the preset cost threshold, the insured person’s In the current hospitalization data, there is an abnormal situation of short hospitalization for examination. Further, if the number of admission days is less than the preset number of days, and the cost of the inpatient examination is less than the preset cost threshold, you can also determine whether there is a follow-up operation or transfer record within the preset time period after admission Determine whether there is any abnormality in the data. If there is no follow-up operation or transfer record within the preset period of time after the insured person is admitted to the hospital, the abnormality of the short hospitalization in the hospitalization data is also determined.

Further, if there is abnormal data for the short hospitalization for examination, the basis for the fee deduction is determined based on the details of the short hospitalization for examination in the abnormal data. It is understandable that the number of admission days is less than the preset threshold and the subsequent preset time period If there is no follow-up operation or transfer record, and the cost of in-patient examination is greater than the regular in-patient examination fee, different deduction rules can be corresponded, and the details of the short-term hospitalization of the violation will be determined to each person who violates the rule, so that Issue notice of deduction.

In this embodiment, it is judged whether the standard data needs to be reduced in dimension, and after it is determined that the dimension reduction is needed, a linear method is used to reduce the dimension of the standard data, so that the standard data after the dimension reduction is based on the clustering algorithm Perform cluster analysis to identify abnormal data including short hospitalization examinations, and send corresponding deduction notices to specific designated medical institutions to manage the outpatient co-ordination fund payment to prevent unreasonable short hospitalization examination expenses.

In addition, an embodiment of the present application also provides a hospitalization data abnormality detection device.

Referring to FIG. 5, FIG. 5 is a schematic diagram of functional modules of the first embodiment of the hospitalization data abnormality detection device of the present application.

In this embodiment, the device for detecting abnormal hospitalization data includes:

Inpatient data acquisition module 10, used to obtain inpatient data of insured persons;

The abnormal data judging module 20 is configured to judge whether there is abnormal data for short hospitalization for examination based on a preset model;

The deduction basis determination module 30 is used to determine the deduction basis based on the inspection details of the short hospitalization in the abnormal data if it is so as to send a deduction notice to the corresponding designated medical institution based on the deduction basis.

Further, the hospitalization data acquisition module 10 specifically includes:

The data acquisition unit is used to obtain the inpatient data of the insured persons uploaded by the designated medical institution, wherein the inpatient data includes at least the designated medical institution, the information of the insured person, the reason for hospitalization, the number of days admitted to the hospital, follow-up operation, Hospital transfer records and hospitalization examination costs.

Further, the abnormal data judgment module 20 specifically includes:

A data conversion unit, used to convert the text data in the hospitalization data into standard data;

The cluster analysis unit is configured to perform cluster analysis on the standard data based on a clustering algorithm, and determine whether there is abnormal data for short hospitalization examination in the standard data based on the result of the cluster analysis.

Further, the data conversion unit specifically includes:

A data cleaning subunit, used for cleaning the hospitalization data to obtain standardized data after cleaning;

The text feature construction subunit is used to construct text features based on the specification data to obtain corresponding word vectors;

The standard data determination subunit is used to encode the word vector into a vector matrix based on a bidirectional recurrent neural network RNN model, so as to determine corresponding standard data based on the vector matrix.

Further, the hospitalization data abnormality detection device further includes:

The dimension reduction judgment unit is used to judge whether the standard data needs to be dimension reduced.

The dimension reduction processing unit is used for performing dimension reduction processing on the standard data based on a linear method, and performing clustering analysis on the standard data based on the clustering algorithm, and based on the result of the clustering analysis The step of judging whether there is abnormal data of short hospitalization for examination in the standard data.

Further, the cluster analysis unit specifically includes:

The cluster analysis subunit is used to perform cluster analysis on the standard data based on a clustering algorithm, and divide the standard data into different categories, where the standard data includes at least the number of hospital admission days and the cost of hospitalization examination;

The abnormal time judgment subunit is used for judging whether there is standard data of the number of admission days less than a preset threshold of days in the different categories;

The abnormal cost judging subunit is used for judging whether the hospitalization examination cost corresponding to the standard data that the number of admission days is less than the preset day threshold is greater than the preset cost threshold.

Further, the deduction basis determination module 30 specifically includes:

A deduction unit, used to obtain a preset deduction rule, and based on the preset deduction rule, and the difference between the number of days in the abnormal data and the threshold of the preset number of days, and the cost of the inpatient examination The cost difference with the preset cost threshold determines the basis for deduction;

The deduction notification unit is used to generate a deduction notification based on the deduction basis, and send the deduction notification to the corresponding designated medical institution.

Wherein, each module in the above-mentioned hospitalization data abnormality detection device corresponds to each step in the above-mentioned hospitalization data abnormality detection method embodiment, and its function and implementation process will not be repeated here one by one.

In addition, an embodiment of the present application further proposes a readable storage medium. The readable storage medium may be a non-volatile readable storage medium. The readable storage medium stores computer-readable instructions. The computer may When the read instruction is executed by the processor, the steps of the hospitalization data abnormality detection method described above are realized.

For the method implemented when the computer-readable instructions are executed, reference may be made to the various embodiments of the hospitalization data abnormality detection method of the present application, which will not be repeated here.

The solution provided in this embodiment first obtains the inpatient data of the insured person, and determines whether there is abnormal data for short hospitalization in the hospitalization data based on a preset model, and if so, based on the short hospitalization in the abnormal data for inspection Determine the basis for deductions in detail so as to send the deduction notice to the corresponding designated medical institution based on the basis for deductions. Through the hospitalization data anomaly detection method proposed in this application, the insured personnel's hospitalization inspection is tested to determine whether there is abnormal data for short hospitalization examinations, and deductions are issued to designated medical institutions that have short hospitalization examination abnormal data Notice to prevent unreasonable outpatient expenses, effectively manage the outpatient co-ordination fund expenditure, and avoid wasting the outpatient co-ordination fund.

It should be noted that in this article, the terms "include", "include" or any other variant thereof are intended to cover non-exclusive inclusion, so that a process, method, article or system that includes a series of elements includes not only those elements, It also includes other elements that are not explicitly listed, or include elements inherent to this process, method, article, or system. Without more restrictions, the element defined by the sentence "include one..." does not exclude that there are other identical elements in the process, method, article or system that includes the element.

The sequence numbers of the above embodiments of the present application are for description only, and do not represent the advantages and disadvantages of the embodiments.

Through the description of the above embodiments, those skilled in the art can clearly understand that the methods in the above embodiments can be implemented by means of software plus a necessary general hardware platform, and of course, can also be implemented by hardware, but in many cases the former is better Implementation. Based on this understanding, the technical solution of the present application can be embodied in the form of a software product in essence or part that contributes to the existing technology, and the computer software product is stored in a storage medium (such as ROM/RAM as described above) , Magnetic disk, optical disk), including several instructions to make a terminal device (which can be a mobile phone, computer, server, air conditioner, or network equipment, etc.) to perform the method described in each embodiment of the present application.

The above are only the preferred embodiments of the present application, and do not limit the scope of the patent of the present application. Any equivalent structure or equivalent process transformation made by the description and drawings of this application, or directly or indirectly used in other related technical fields , The same reason is included in the scope of patent protection in this application.

Claims

A method for detecting abnormal hospitalization data, wherein the method for detecting hospitalization includes the following steps:

Obtain medical data of insured persons;

Determine whether there is abnormal data for short hospitalization for examination based on a preset model;

If yes, determine the basis for the fee deduction based on the details of the short hospital stay in the abnormal data, so as to send a fee deduction notice to the corresponding designated medical institution based on the basis for the fee deduction;

Wherein, the step of determining whether there is abnormal data for short hospitalization for examination based on the preset model includes:

Convert the text data in the hospitalization data into standard data;

Perform cluster analysis on the standard data based on a clustering algorithm, and determine whether there is abnormal data in the standard data for short hospitalization examination based on the result of the cluster analysis. .
The abnormality detection method for hospitalization data according to claim 1, wherein the step of obtaining the hospitalization data of the insured personnel includes:

Obtain the hospitalization data of the insured personnel uploaded by the designated medical institution, where the hospitalization data at least includes the designated medical institution, information of the insured person, the reason for hospitalization, the number of days admitted, surgery follow-up, transfer records and hospitalization examination costs .
The abnormality detection method for hospitalization data according to claim 1, wherein the step of converting the text data in the hospitalization data into standard data includes:

Clean the hospitalization data to obtain the standardized data after cleaning; construct text features based on the standardized data to obtain the corresponding word vectors;

The word vector is encoded into a vector matrix based on a bidirectional recurrent neural network RNN model, so as to determine corresponding standard data based on the vector matrix.
The abnormality detection method for hospitalization data according to claim 1, wherein after the step of converting the text data in the hospitalization data into standard data, the method further comprises:

Determine whether the standard data needs to be reduced in dimension;

If yes, perform dimensionality reduction processing on the standard data based on the linear method, and perform cluster analysis on the standard data based on the clustering algorithm, and determine whether the standard data is based on the result of the cluster analysis There is a procedure of abnormal data for short hospitalization for examination.
The abnormality detection method for hospitalization data according to claim 1, wherein the clustering algorithm performs cluster analysis on the standard data and determines whether there is a short hospitalization in the standard data based on the result of the cluster analysis The steps to check the abnormal data include:

Perform cluster analysis on the standard data based on a clustering algorithm, and divide the standard data into different categories, where the standard data includes at least the number of days of admission and the cost of hospitalization examination;

Judging whether there is standard data for the number of admission days less than the preset threshold in the different categories;

If yes, it is judged whether the inpatient examination cost corresponding to the standard data of the number of admission days is less than the preset number of days threshold is greater than the preset cost threshold.
The abnormality detection method for hospitalization data according to claim 5, wherein the basis for determining the deduction basis based on the inspection details of the short hospitalization in the abnormality data is to send a deduction to the corresponding designated medical institution based on the deduction basis The fee notification steps include:

Obtain a preset fee deduction rule, and based on the preset fee deduction rule, and the difference in the number of days between the admission days and the preset number of days threshold in the abnormal data, and the The cost difference determines the basis for deduction;

Generate a deduction notice based on the deduction basis, and send the deduction notice to the corresponding designated medical institution.
A hospitalization data abnormality detection device, wherein the hospitalization data abnormality detection device includes:

The hospitalization data acquisition module is used to obtain the hospitalization data of the insured personnel;

The abnormal data judgment module is used to judge whether there is abnormal data for short hospitalization for examination based on a preset model;

The deduction basis determination module is used to determine the deduction basis based on the inspection details of the short hospitalization in the abnormal data if it is so as to send a deduction notice to the corresponding designated medical institution based on the deduction basis;

Wherein, the abnormal data judgment module includes:

A data conversion unit, used to convert the text data in the hospitalization data into standard data;

The cluster analysis unit is configured to perform cluster analysis on the standard data based on a clustering algorithm, and determine whether there is abnormal data for short hospitalization examination in the standard data based on the result of the cluster analysis.
The hospitalization data abnormality detection device according to claim 7, wherein the hospitalization data acquisition module includes:

The data acquisition unit is used to obtain the inpatient data of the insured persons uploaded by the designated medical institution, wherein the inpatient data includes at least the designated medical institution, the information of the insured person, the reason for hospitalization, the number of days admitted to the hospital, follow-up operation, Hospital transfer records and hospitalization examination costs.
The hospitalization data abnormality detection device according to claim 7, wherein the data conversion unit includes:

A data cleaning subunit, used for cleaning the hospitalization data to obtain standardized data after cleaning;

The text feature construction subunit is used to construct text features based on the specification data to obtain corresponding word vectors;

The standard data determination subunit is used to encode the word vector into a vector matrix based on a bidirectional recurrent neural network RNN model, so as to determine corresponding standard data based on the vector matrix.
The hospitalization data abnormality detection device according to claim 7, wherein the hospitalization data abnormality detection device further comprises:

The dimension reduction judgment unit is used to judge whether the standard data needs to be dimension reduced.

The dimensionality reduction processing unit is used for performing dimensionality reduction processing on the standard data based on a linear method, performing cluster analysis on the standard data based on a clustering algorithm, and judging based on the result of the cluster analysis Whether there is abnormal data in the standard data for short hospitalization.
The hospitalization data abnormality detection device according to claim 7, wherein the cluster analysis unit specifically includes:

The cluster analysis subunit is used to perform cluster analysis on the standard data based on a clustering algorithm, and divide the standard data into different categories, where the standard data includes at least the number of hospital admission days and the cost of hospitalization examination;

The abnormal time judgment subunit is used for judging whether there is standard data of the number of admission days less than a preset threshold of days in the different categories;

The abnormal cost judging subunit is used for judging whether the hospitalization examination cost corresponding to the standard data that the number of admission days is less than the preset day threshold is greater than the preset cost threshold.
The hospitalization data abnormality detection device according to claim 11, wherein the deduction basis determination module specifically includes:

A deduction unit, used to obtain a preset deduction rule, and based on the preset deduction rule, and the difference between the number of days in the abnormal data and the threshold of the preset number of days, and the cost of the inpatient examination The cost difference with the preset cost threshold determines the basis for deduction;

The deduction notification unit is used to generate a deduction notification based on the deduction basis, and send the deduction notification to the corresponding designated medical institution.
A hospitalization data abnormality detection device, wherein the hospitalization data abnormality detection device includes: a memory, a processor, and computer-readable instructions stored on the memory and operable on the processor, the computer-readable When the instruction is executed by the processor, the following steps are realized:

Obtain medical data of insured persons;

Determine whether there is abnormal data for short hospitalization for examination based on a preset model;

If yes, determine the basis for the fee deduction based on the details of the short hospital stay in the abnormal data, so as to send a fee deduction notice to the corresponding designated medical institution based on the basis for the fee deduction;

When the computer-readable instructions are executed by the processor, the following steps are implemented:

Convert the text data in the hospitalization data into standard data;

Perform cluster analysis on the standard data based on a clustering algorithm, and determine whether there is abnormal data in the standard data for short hospitalization examination based on the result of the cluster analysis.
The hospitalization data abnormality detection device according to claim 13, wherein when the computer-readable instructions are executed by the processor, the following steps are further implemented:

Clean the hospitalization data to obtain the standardized data after cleaning; construct text features based on the standardized data to obtain the corresponding word vectors;

The word vector is encoded into a vector matrix based on a bidirectional recurrent neural network RNN model, so as to determine corresponding standard data based on the vector matrix.
The hospitalization data abnormality detection device according to claim 13, wherein when the computer-readable instructions are executed by the processor, the following steps are further implemented:

Determine whether the standard data needs to be reduced in dimension;

If yes, perform dimensionality reduction processing on the standard data based on the linear method, and perform cluster analysis on the standard data based on the clustering algorithm, and determine whether the standard data is based on the result of the cluster analysis There is a procedure of abnormal data for short hospitalization for examination.
The hospitalization data abnormality detection device according to claim 13, wherein when the computer-readable instructions are executed by the processor, the following steps are further implemented:

Perform cluster analysis on the standard data based on a clustering algorithm, and divide the standard data into different categories, where the standard data includes at least the number of days of admission and the cost of hospitalization examination;

Judging whether there is standard data for the number of admission days less than the preset threshold in the different categories;

If yes, it is judged whether the inpatient examination cost corresponding to the standard data of the number of admission days is less than the preset number of days threshold is greater than the preset cost threshold.
A readable storage medium, wherein computer readable instructions are stored on the readable storage medium, and when the computer readable instructions are executed by a processor, the following steps are implemented:

Obtain medical data of insured persons;

Determine whether there is abnormal data for short hospitalization for examination based on a preset model;

If yes, determine the basis for the fee deduction based on the details of the short hospital stay in the abnormal data, so as to send a fee deduction notice to the corresponding designated medical institution based on the basis for the fee deduction;

When the computer readable instructions are executed by the processor, the following steps are realized

Convert the text data in the hospitalization data into standard data;

Perform cluster analysis on the standard data based on a clustering algorithm, and determine whether there is abnormal data in the standard data for short hospitalization examination based on the result of the cluster analysis.
The readable storage medium of claim 17, wherein the computer readable instructions when executed by the processor further implement the following steps:

Clean the hospitalization data to obtain the standardized data after cleaning; construct text features based on the standardized data to obtain the corresponding word vectors;

The word vector is encoded into a vector matrix based on a bidirectional recurrent neural network RNN model, so as to determine corresponding standard data based on the vector matrix.
The readable storage medium of claim 17, wherein the computer readable instructions when executed by the processor further implement the following steps:

Determine whether the standard data needs to be reduced in dimension;

If yes, perform dimensionality reduction processing on the standard data based on the linear method, and perform cluster analysis on the standard data based on the clustering algorithm, and determine whether the standard data is based on the result of the cluster analysis There is a procedure of abnormal data for short hospitalization for examination.
The readable storage medium of claim 17, wherein the computer readable instructions when executed by the processor further implement the following steps:

Perform cluster analysis on the standard data based on a clustering algorithm, and divide the standard data into different categories, where the standard data includes at least the number of days of admission and the cost of hospitalization examination;

Judging whether there is standard data for the number of admission days less than the preset threshold in the different categories;

If yes, it is judged whether the inpatient examination cost corresponding to the standard data of the number of admission days is less than the preset number of days threshold is greater than the preset cost threshold.