CN115700683A

CN115700683A - Medical rationality determination method, apparatus and program product

Info

Publication number: CN115700683A
Application number: CN202110859366.4A
Authority: CN
Inventors: 杨倩文
Original assignee: Alibaba Singapore Holdings Pte Ltd
Current assignee: Alibaba Innovation Co
Priority date: 2021-07-28
Filing date: 2021-07-28
Publication date: 2023-02-07

Abstract

The embodiment of the application provides a method, equipment and a program product for determining medical rationality. In the embodiment of the application, the medical insurance data of the patient to be audited can be subjected to data analysis, and the medical action path of the patient to be audited is determined; the medical behavior path of the patient is subjected to feature extraction, and medical rationality analysis is performed according to the extracted medical behavior features of the patient to be examined so as to determine the medical rationality of the patient to be examined, so that the medical rationality can be automatically determined, and the efficiency of identifying medical fraudulent behaviors is improved.

Description

Medical rationality determination method, apparatus and program product

Technical Field

The present application relates to the field of data processing technologies, and in particular, to a method, a device, and a program product for determining medical rationality.

Background

In recent years, medical fraud has received much attention. Some medical institutions, doctor diagnosis data counterfeiting, and through intensive care of small diseases, excessive medical cheating to obtain medical insurance and patient funds, serious influence is caused on the medical insurance order. Therefore, how to effectively identify medical fraud becomes an urgent problem to be solved in the technical field of data processing.

Disclosure of Invention

Aspects of the present application provide a method, device, and program product for determining medical rationality, which are used to automatically determine medical rationality and contribute to improving the efficiency of identifying medical fraudulent activities.

The embodiment of the application provides a method for determining medical rationality, which comprises the following steps:

acquiring medical insurance data of a patient to be audited;

performing data analysis on the medical insurance data to determine a medical action path of the patient to be audited;

performing feature extraction on the medical behavior path to obtain medical behavior features of the patient to be audited;

and performing medical rationality analysis according to the medical behavior characteristics to determine the medical rationality of the patient to be examined.

The embodiment of the application further provides a method for determining medical rationality, which comprises the following steps:

responding to the medical rationality analysis request, and acquiring medical insurance data of the patient to be audited;

performing medical rationality analysis according to the medical behavior characteristics to determine the medical rationality of the patient to be examined;

and providing the medical rationality to a client sending out the rationality analysis request so as to enable the client to output the medical rationality.

An embodiment of the present application further provides a computer device, including: a memory and a processor; wherein the memory is used for storing a computer program;

the processor is coupled to the memory for executing the computer program for performing the steps of the medical rationality determination method described above.

An embodiment of the present application further provides a computer program product, including a computer program; the medical rationality determination method may be implemented when the computer program is executed by one or more processors.

Embodiments of the present application also provide a computer-readable storage medium storing computer instructions that, when executed by one or more processors, cause the one or more processors to perform the steps in the medical rationality method described above.

In the embodiment of the application, the medical insurance data of the patient to be audited can be subjected to data analysis, and the medical action path of the patient to be audited is determined; the medical behavior path of the patient is subjected to feature extraction, and medical rationality analysis is performed according to the extracted medical behavior features of the patient to be examined so as to determine the medical rationality of the patient to be examined, so that the medical rationality can be automatically determined, and the efficiency of identifying medical fraudulent behaviors can be improved.

Drawings

The accompanying drawings, which are included to provide a further understanding of the application and are incorporated in and constitute a part of this application, illustrate embodiment(s) of the application and together with the description serve to explain the application and not to limit the application. In the drawings:

FIG. 1 is a block diagram of a data processing system according to an embodiment of the present application;

FIG. 2 is a schematic diagram of an improved RNN model according to an embodiment of the present application;

fig. 3 and 4 are schematic flow charts of a medical rationality determination method provided by an embodiment of the present application;

fig. 5 is a schematic structural diagram of a computer device according to an embodiment of the present application.

Detailed Description

In order to make the objects, technical solutions and advantages of the present application more apparent, the technical solutions of the present application will be described in detail and completely with reference to the following specific embodiments of the present application and the accompanying drawings. It should be apparent that the described embodiments are only some of the embodiments of the present application, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present application.

Medical disorder sometimes occurs. Some medical institutions, doctor diagnosis data counterfeiting, and through intensive care of small diseases, excessive medical cheating to obtain medical insurance and patient funds, serious influence is caused on the medical insurance order.

In order to effectively identify medical fraud, in some embodiments of the present application, data analysis may be performed on medical insurance data of a patient to be audited, and a medical behavior path of the patient to be audited is determined; the medical behavior path of the patient is subjected to feature extraction, and medical rationality analysis is performed according to the extracted medical behavior features of the patient to be examined so as to determine the medical rationality of the patient to be examined, so that the medical rationality can be automatically determined, and the efficiency of identifying medical fraudulent behaviors is improved.

The technical solutions provided by the embodiments of the present application are described in detail below with reference to the accompanying drawings.

It should be noted that: like reference numerals refer to like objects in the following figures and embodiments, and thus, once an object is defined in one figure or embodiment, further discussion thereof is not required in subsequent figures and embodiments.

Fig. 1 is a schematic structural diagram of a data processing system according to an embodiment of the present application. As shown in fig. 1, the system includes: a terminal device 11 and a server device 12.

Wherein, the service end device 12 and the terminal device 11 may be connected wirelessly or by wire. Optionally, the server device 12 may be communicatively connected to the terminal device 11 through a mobile network, and accordingly, the network format of the mobile network may be any one of 2G (GSM), 2.5G (GPRS), 3G (WCDMA, TD-SCDMA, CDMA2000, UTMS), 4G (LTE), 4G + (LTE +), 5G, wiMax, and the like. Alternatively, the server device 12 may also be communicatively connected to the terminal device 11 through bluetooth, wiFi, infrared, or the like.

In this embodiment, the terminal device 11 refers to a computer device used by a user and having functions of computing, accessing internet, communicating and the like required by the user, and may be, for example, a smart phone, a tablet computer, a personal computer, a wearable device and the like. In this embodiment, the user may be a medical staff, a patient, or a medical staff in charge of medical insurance, etc.

In this embodiment, the server device 12 is a computer device that can manage medical insurance data and respond to a service request from the terminal device 11 to provide a service related to data processing for a user, and generally has the capability of undertaking and guaranteeing the service. The server device 12 may be a single server device, a cloud server array, or a Virtual Machine (VM) running in the cloud server array. In addition, the server device may also refer to other computing devices with corresponding service capabilities, such as a terminal device (running a service program) such as a computer.

The server device 12 may be a server of a medical insurance system or a server of a hospital. Alternatively, the server device 12 may be a server in a medical insurance system or a private cloud of a hospital. Alternatively, the server device 12 may also be a server in a public cloud, and the like.

In this embodiment, the terminal device 11 may send a medical rationality analysis request to the server device 12 in response to the audit event. Accordingly, the server device 12 may receive a medical rationality analysis request; and responding to the medical rationality analysis request to acquire medical insurance data of the patient to be audited. The medical insurance data refers to data of a patient for medical reimbursement, and may include but is not limited to: a medical record home page, a medicine list, a medical examination item list and the like. Wherein, the first page of the medical record records the personal information of the patient, the outpatient diagnosis, the admission diagnosis and the like. The list of medications may include: name of medicine, price, quantity, and time to prescribe a drug, etc. The medicine includes medicine, medical consumables, etc. The medical examination item list includes: name, price, examination time, and the like of the medical examination item.

In the embodiment of the present application, a specific implementation form of the audit event is not limited. In some embodiments, as shown in fig. 1, the terminal device 11 may provide a medical analysis interface K1. Alternatively, the medical analysis interface may be implemented as the medical analysis control K1. Accordingly, the audit event may be implemented as an audit event generated for a triggering operation of the medical analysis interface K1.

In the embodiment of the present application, a specific implementation manner of the server device 12 obtaining medical insurance data of a patient to be audited is not limited. In some embodiments, the medical insurance data can be uploaded to the medical insurance system by medical insurance responsible personnel of the medical institution through terminal equipment of the medical insurance responsible personnel; and performing medical insurance accounting by the medical insurance system. The medical insurance can be a real-time reimbursement medical insurance, such as rural medical care, social insurance and the like. Based on this, in this embodiment, the server device 12 may obtain the identification information of the patient to be audited from the medical rationality analysis request; and acquiring medical insurance data corresponding to the identification information of the patient to be audited from server-side equipment on the medical insurance system side, wherein the medical insurance data is used as the medical insurance data of the patient to be audited. The patient identification information refers to information that can uniquely identify a patient. Such as, but not limited to, the patient identifier may be the patient's name, identification number, social security number, etc. The server device 12 and the server device on the medical insurance system side may be the same device or different server devices.

In other embodiments, the medical insurance data is a non-real-time reimbursement medical insurance, such as a commercial insurance. The user can upload medical insurance data of a patient to be audited to the server-side device 12 through the terminal device 11. Optionally, as shown in fig. 1, the terminal device 11 may provide a data uploading interface, such as a medical insurance data providing control K2. The user can upload the medical insurance data through the medical insurance data providing control. The terminal device 11 may respond to the trigger operation of the medical insurance data providing control K2, and obtain data associated with the trigger operation as medical insurance data of the patient to be audited. Namely, the data uploaded by the terminal device 11 through the medical insurance data providing control is used as the medical insurance data of the patient to be audited. In this embodiment, the terminal device 11 may further provide medical insurance data of the patient to be audited to the server device 12. Optionally, the terminal device 11 may provide the medical insurance data of the patient to be audited to the server device 12 before initiating the medical rationality analysis request, store the medical insurance data of the patient by the server device 12, and establish a corresponding relationship between the patient identification information and the medical insurance data of the patient.

Further, when the user audits the medical rationality of the patient, the medical audit interface K1 can be triggered. The terminal device 11 may issue a medical rationality analysis request to the server device 12 in response to an audit event generated for triggering the medical audit interface K1. The medical rationality analysis request may include identification information of the patient. Optionally, the terminal device 11 may provide a patient identification upload interface K3. Optionally, the terminal device 11 may provide a patient identification input control through which the user may input identification information of the patient to be audited. In other embodiments, as shown in fig. 1, the patient identifier uploading interface K3 may also be implemented as a selection control K3, through which the user may select the identifier of the patient to be audited. Accordingly, the terminal device 11 may, in response to the selection operation for the patient identifier, use the selected patient identifier as the identifier of the patient to be audited.

The server-side device 12 may respond to the medical rationality analysis request to obtain medical insurance data corresponding to the patient identification information as medical data of the patient to be audited. Optionally, the server device 12 may analyze the patient identification information from the medical rationality analysis request, and match the analyzed patient identification information with the corresponding relationship between the patient identification information and the medical insurance data of the patient to obtain the medical insurance data corresponding to the analyzed patient identification information, which is used as the medical insurance data of the patient to be audited.

After obtaining the medical insurance data of the patient to be audited, the server device 12 may perform data analysis on the medical insurance data to determine the medical action path of the patient to be audited. In some embodiments, the medical insurance data includes: image data. The server device 12 may perform text recognition on the image data to extract text data included in the image data; further, the server device 12 may perform data analysis on the text data to determine a medical action path of the patient to be audited. Alternatively, server device 12 may perform text Recognition on the image data by using an Optical Character Recognition (OCR) technology to extract text data included in the image data. For the text medical insurance data, the server-side device 12 can directly perform data analysis on the text medical insurance data to determine the medical behavior data of the patient to be audited.

Optionally, the server device 12 may perform text analysis on the medical insurance data of the patient to be audited to determine the medical entity included in the medical insurance data. The medical entity refers to an entity word which can reflect the information of the medical item. The medical item information may include: medicine information, medical examination item information, treatment item information, and the like. The medical information may include: drug name, quantity, manufacturer information, time to prescribe a drug, etc. Such as amoxicillin, 3 boxes, a pharmaceutical company, no. 3/8/2020. The medical examination item information may include: the item name and the check time are checked. Wherein the medical examination item may include: blood routine, urine routine, computed Tomography (CT), magnetic resonance examination, and the like. The treatment item information may include: the name of the treatment item (such as laparoscopic surgery), the implementation time of the treatment item, and the like.

In some embodiments, the server device 12 may perform semantic analysis on medical insurance data of a patient to be audited to obtain keywords from the medical insurance data. Optionally, the server-side device 12 may perform word segmentation processing on the medical insurance data to obtain a word set included in the medical insurance data. Optionally, the server device 12 may perform word segmentation processing on the medical insurance data by using word vector models such as a deformer-based Bidirectional Encoding Representation (BERT) model, a word2vec model, an embedded language model (ELMo model), or a generated forward training (GPT) model, to obtain a word set included in the medical insurance data.

Further, the server-side device 12 may perform attribute identification on the word set to determine attributes of words in the word set; and then determining keywords reflecting the medical entity according to the attributes of all the words in the word set, wherein the keywords are used as the medical entity included in the medical insurance data. Optionally, the keywords with the word attributes as the target attributes may be selected according to the attributes of each word in the word set. Wherein the target attribute may be a descriptive item of the medical item information. Such as treatment item names, medicine names, medical examination item names, and the like.

Further, the server device 12 may determine a medical action corresponding to the medical entity according to the attribute of the medical entity. For example, the medical action of the medical examination item may be determined as a medical examination; the medical action of the medicine is prescription and the like; the medical action of the treatment item is drug treatment or surgical treatment.

Further, the server device 12 may sequence the medical actions according to the occurrence sequence of the medical actions, so as to obtain a medical action path of the patient to be audited. The medical action path of the patient to be audited can be implemented as serialized medical item information, such as a medical item information sequence arranged according to the occurrence sequence of the medical actions. Such as blood routine examination information- > liver function examination information- > CT information- > color Doppler ultrasound information- > medicine B information (medicine for treating hepatitis), etc.

Based on the consideration that the medical action path of the patient can reflect the sequential characteristics of the medical actions and the medical item information included in the medical action path can reflect the characteristics of the medical item, the server-side device 12 can perform characteristic extraction on the medical action path to obtain the medical action characteristics of the patient to be audited.

Optionally, the server device 12 may input the medical action path of the patient to be audited into the network model; in the network model, feature extraction is carried out on the medical behavior path to obtain the medical behavior features of the patient to be examined. In the embodiment of the present application, a concrete implementation form of the network model is not limited. Alternatively, the network model may be implemented as a multiple-input multiple-output network model; such as a Recurrent Neural Network (RNN) model or a long-short-term memory (LSTM) model; of course, other forms of network models may be implemented, such as Convolutional Neural Network (CNN) models, deep Neural Network (DNN) models, or decision tree models, among others.

Considering that the medical action paths of different disease types are different, a corresponding network model can be trained for each disease type separately to extract the features of the medical action paths of the disease type. For convenience of description and distinction, in the embodiment of the present application, a network model for performing feature extraction on a medical action path is collectively referred to as a feature extraction model; and defining the network model for extracting the characteristics of the medical action path of the patient to be examined as a first network model. Based on the above implementation mode that each disease category corresponds to one feature extraction model, in this embodiment, the disease category of the patient to be audited can be obtained from the medical insurance data of the patient to be audited; and acquiring a feature extraction model corresponding to the disease type of the patient to be examined from the feature extraction model as a first network model for performing feature extraction on the medical behavior feature of the patient to be examined.

Aiming at the multi-input multi-output network model, the medical entity information corresponding to the medical action path can be sequentially input into a plurality of sub-models of the network model according to the occurrence sequence of the medical actions; in the network model, aiming at two adjacent submodels, the next submodel can be used for carrying out feature extraction on the medical behavior feature output by the previous submodel and the medical entity information input to the next submodel so as to obtain the medical behavior feature output by the next submodel; and determining the medical behavior characteristics output by the plurality of sub-models as the medical behavior characteristics of the patient to be examined. For example, in the multi-input multi-output RNN model shown in fig. 2, x1-xk respectively represent k pieces of medical entity information input into a plurality of sub models, and y1-yk respectively represent medical behavior characteristics output by the plurality of sub models, that is, medical behavior characteristics of a patient to be examined.

Regardless of the form in which the first network model is implemented, the first network model is trained before the first network model is used to perform feature extraction on the patient behavior path. In this embodiment, in order to implement training of the first network model, a reasonable medical behavior path may be determined according to medical insurance data with a known reasonable medical behavior; unreasonable medical behaviors are added in the reasonable medical behavior paths to obtain unreasonable medical behavior paths; and then, performing model training by taking the reasonable medical behavior path as a positive sample and taking the unreasonable medical behavior path as a negative sample to obtain a first network model.

Aiming at the embodiment that different disease species correspond to different feature extraction models, the reasonable medical action path can be determined by adopting medical insurance data corresponding to the known medical action under the disease species; unreasonable medical behaviors are added in the reasonable medical behavior paths to obtain unreasonable medical behavior paths; and then, performing model training by taking the reasonable medical behavior path as a positive sample and taking the unreasonable medical behavior path as a negative sample to obtain a feature extraction model corresponding to the disease species.

After the medical behavior characteristics of the patient to be audited are obtained, the server-side device 12 may perform reasonability analysis according to the medical behavior characteristics to determine the medical reasonability of the patient to be audited. Wherein the medical rationality includes medical rationality and medical irrationality.

Optionally, the server device 12 may input the medical behavior characteristics of the patient to be audited into the second network model, and in the second network model, calculate the medical rationality probability of the patient to be audited according to the medical behavior characteristics of the patient to be audited; and then, determining the rationality of the patient to be examined according to the medical rationality probability of the patient to be examined. Optionally, the reasonable and unreasonable probability of the medical behavior of the patient to be audited can be calculated according to the medical behavior characteristics of the patient to be audited; and determining the rationality of reasonable medical behaviors and high probability of incoordination as the medical rationality of the patient to be examined.

Wherein the second network model may be implemented as a classifier of the first network model, such as an activation function layer of the first network model, a decision tree model or other neural network model, and so on. The activation function may be a softmax function, a tanh function, a sigmoid function, a Relu function, or the like.

In the embodiment of the application, the second network model can be obtained by training according to medical behavior characteristics obtained by medical insurance data of a single disease category. Because the medical behavior characteristics of different disease categories are different, when the second network model obtained by training the medical behavior characteristics of a single disease category is used for reasonably analyzing the similar disease category of the disease category and the medical behavior characteristics of multiple complications, errors may be caused in the result of reasonable analysis. Based on this, in the embodiment of the present application, the medical rationality analysis model may also be trained for the similar disease types and various complications, and defined as a third network model. Alternatively, the third network model may be implemented as a complex classifier, such as a fuzzy neural network or the like.

The third network model can perform model training for training samples according to medical insurance data of similar disease types and medical insurance characteristics obtained by medical insurance data of combined treatment of multiple complications. Specifically, medical insurance data of a plurality of disease types of which known disease types are similar disease types and medical insurance data of combined treatment of a plurality of complications can be obtained; performing data analysis on medical insurance data of similar disease species and medical insurance data of multiple complication combined treatments to determine medical behavior paths of the similar disease species and medical behavior paths of the multiple complication combined treatments; and reasonably marking the medical action paths of similar disease species and the medical action paths of multiple complication combined treatments in a manual marking mode, and acquiring the medical action paths of the similar disease species with reasonable known medical actions and the medical action paths of the multiple complication combined treatments with reasonable medical actions.

Further, medical behavior paths of similar disease categories with reasonable known medical behaviors are respectively input into the first network models corresponding to the corresponding disease categories, and the medical behavior paths of similar medical insurance with reasonable known medical behaviors are subjected to feature extraction in the first network models to obtain reasonable medical behavior features of the similar disease categories; and inputting the medical behavior characteristics of the known medical behavior reasonable multiple complication combined treatment into a first network model corresponding to the corresponding disease category, and performing characteristic extraction on the medical behavior paths of the multiple complication combined treatment in the first network model to obtain the reasonable medical behavior characteristics of the multiple complication combined treatment. Further, the reasonable medical behavior characteristics of similar disease species and the reasonable medical behavior characteristics of combined treatment of multiple complications can be used as training samples for model training to obtain a third network model.

Further, in order to improve the accuracy of the medical rationality analysis result and reduce the probability of errors in the medical rationality analysis caused by the combined treatment of similar diseases or multiple complications, in this embodiment, in view of the above-mentioned medical rationality using the second network model being not medically reasonable, the medical behavior characteristics of the patient to be examined may be input into the third network model; in the third network model, medical rationality analysis can be performed according to the medical behavior characteristics to obtain the medical rationality of the patient to be audited. Because the third network model is obtained according to the medical insurance data of similar disease types and the medical insurance data of combined treatment of multiple complications, the third network model is used for medical rationality analysis, the fuzzy membership degree of the similar disease types and the complications can be improved, and the error rate of rationality analysis and classification can be reduced.

Further, as shown in fig. 1, the server-side device 12 may also provide the medical rationality (i.e., the medical rationality analysis result) of the patient to be audited to the terminal device 11. Accordingly, the terminal device 11 may receive the medical rationality of the patient to be audited and output the medical rationality of the patient to be audited. Optionally, the terminal device 11 may display the medical rationality of the patient to be audited on a screen; and/or playing the medical rationality of the patient to be examined and the like through an audio component.

The data processing system provided by the embodiment can perform data analysis on medical insurance data of a patient to be audited, and determine a medical action path of the patient to be audited; the medical behavior path of the patient is subjected to feature extraction, and medical rationality analysis is performed according to the extracted medical behavior features of the patient to be examined so as to determine the medical rationality of the patient to be examined, so that the medical rationality can be automatically determined, and the efficiency of identifying medical fraudulent behaviors can be improved.

In addition to the above system embodiments, the present application embodiment also provides a disease species identification method, and the disease species identification method provided in the present application embodiment is exemplarily described below.

Fig. 3 is a schematic flow chart of a disease identification method according to an embodiment of the present application. As shown in fig. 3, the method includes:

301. and acquiring medical insurance data of the patient to be audited.

302. And performing data analysis on the medical insurance data to determine the medical action path of the patient to be audited.

303. And performing feature extraction on the medical action path to obtain the medical action features of the patient to be examined.

304. And performing medical rationality analysis according to the medical behavior characteristics to determine the medical rationality of the patient to be audited.

The medical rationality determination method provided by the embodiment can be deployed on any computer device. For example, the method can be deployed on a terminal device of a user, and can also be deployed on a server device. No matter which device the execution subject of the medical rationality determination method provided by the embodiment is, in step 301, medical insurance data of a patient to be audited may be acquired. The description of the medical insurance data can refer to the related contents of the above system embodiments, and is not repeated herein.

For the terminal equipment, a medical insurance data providing control can be provided. The user can upload the medical insurance data through the medical insurance data providing control. For the terminal device, in response to the trigger operation of the medical insurance data providing control, data associated with the trigger operation can be acquired as medical insurance data of the patient to be audited. The data uploaded by the user through the medical insurance data providing control is obtained and used as the medical insurance data of the patient to be audited.

Or the terminal equipment can also read medical insurance data of the patient to be audited from the storage medium. The storage medium may be a hard disk fixedly installed on the terminal device, a cloud storage, or an external storage medium such as a usb disk. Optionally, the terminal device may provide a patient identification selection control, through which the user may determine the patient to be audited. Correspondingly, the terminal device can respond to the triggering operation of the patient identification selection control, and acquire the selected patient identification as the identification information of the patient to be audited. Further, the terminal device can match the identification information of the patient to be audited in the corresponding relation between the stored patient identification and the medical insurance data to determine the medical insurance data of the patient to be audited.

Aiming at the server-side equipment, the identification information of the patient to be audited can be analyzed from the received medical rationality analysis request; and matching the identification information of the patient to be audited in the corresponding relation between the patient identification information and the medical record data to obtain the medical insurance data of the patient to be audited.

After obtaining the medical insurance data of the patient to be audited, then, in step 302, data analysis may be performed on the medical insurance data to determine a medical action path of the patient to be audited. In some embodiments, the medical insurance data includes: image data. Text recognition can be performed on the image data to extract text data contained in the image data; further, data analysis can be performed on the text data to determine the medical action path of the patient to be audited. Alternatively, OCR technology may be employed to perform text recognition on the image data to extract text data contained in the image data. Aiming at the text medical insurance data, the data analysis can be directly carried out on the text medical insurance data, and the medical behavior data of the patient to be audited is determined.

Optionally, the medical insurance data of the patient to be audited may be text analyzed to determine the medical entity included in the medical insurance data. Further, the medical behavior corresponding to the medical entity can be determined according to the attribute of the medical entity. And then, sequencing the medical behaviors according to the occurrence sequence of the medical behaviors to obtain a medical behavior path of the patient to be audited.

Based on the consideration that the medical action path of the patient can reflect the sequential characteristics of the medical actions and the medical item information included in the medical action path can reflect the characteristics of the medical item, in step 303, the medical action path can be subjected to characteristic extraction to obtain the medical action characteristics of the patient to be examined.

Optionally, the medical action path of the patient to be audited can be input into the network model; in the network model, feature extraction is carried out on the medical behavior path to obtain the medical behavior features of the patient to be examined.

Aiming at the multi-input multi-output network model, the medical entity information corresponding to the medical action path can be sequentially input into a plurality of sub-models of the network model according to the occurrence sequence of the medical actions; in the network model, aiming at two adjacent submodels, the next submodel can be used for carrying out feature extraction on the medical behavior feature output by the previous submodel and the medical entity information input to the next submodel so as to obtain the medical behavior feature output by the next submodel; and determining the medical behavior characteristics output by the plurality of submodels as the medical behavior characteristics of the patient to be audited.

Regardless of the form in which the first network model is implemented, the first network model is trained before the first network model is used to perform feature extraction on the patient behavior path. In this embodiment, in order to implement training of the first network model, a reasonable medical behavior path may be determined according to medical insurance data with a known reasonable medical behavior; unreasonable medical behaviors are added in the reasonable medical behavior paths to obtain unreasonable medical behavior paths; and then, performing model training by taking the reasonable medical action path as a positive sample and taking the unreasonable medical action path as a negative sample to obtain a first network model.

After obtaining the medical behavior characteristics of the patient to be audited, in step 304, a rationality analysis may be performed according to the medical behavior characteristics to determine the medical rationality of the patient to be audited. Wherein the medical rationality comprises medical rationality and medical unreasonable.

Optionally, the medical behavior characteristics of the patient to be audited may be input into the second network model, and the medical rationality probability of the patient to be audited is calculated in the second network model according to the medical behavior characteristics of the patient to be audited; and then, determining the rationality of the patient to be examined according to the medical rationality probability of the patient to be examined. Optionally, the reasonable and unreasonable probability of the medical behavior of the patient to be audited can be calculated according to the medical behavior characteristics of the patient to be audited; and determining the reasonableness with higher probability of medical behavior reasonableness and medical behavior incoordination as the medical reasonability of the patient to be audited.

In the embodiment of the application, the second network model can be obtained by training according to medical behavior characteristics obtained from medical insurance data of a single disease category. Because the medical behavior characteristics of different disease categories are different, when the second network model obtained by training the medical behavior characteristics of a single disease category is used for reasonably analyzing the similar disease category of the disease category and the medical behavior characteristics of multiple complications, errors can be caused in the reasonable analysis result. Based on this, in the embodiment of the present application, the medical rationality analysis model may also be trained for similar disease types and multiple complications, and defined as a third network model. For the training process of the third network model, reference may be made to the related contents of the above system embodiments, and details are not repeated here.

Further, in order to improve the accuracy of the medical rationality analysis result and reduce the probability of errors in the medical rationality analysis caused by the combination treatment of similar diseases or multiple complications, in this embodiment, the medical behavior characteristics of the patient to be examined can be input into the third network model in response to the medical rationality of the second network model being not medically reasonable; in the third network model, medical rationality analysis can be performed according to the medical behavior characteristics to obtain the medical rationality of the patient to be audited. Because the third network model is obtained according to the medical insurance data of similar disease types and the medical insurance data of combined treatment of multiple complications, the third network model is used for medical rationality analysis, the fuzzy membership degree of the similar disease types and the complications can be improved, and the error rate of rationality analysis and classification can be reduced.

Further, if the execution subject of the medical rationality determining method is terminal equipment, the medical rationality result (such as medical rationality or unreasonable) of the patient to be examined can be displayed on a screen; and/or, the disease types of the patients are audited through the audio component. If the execution subject of the medical rationality determining method is the server device, the medical rationality result of the patient to be examined can be sent to the client (such as the terminal device) initiating the medical rationality analysis request. Correspondingly, the client can receive the medical rationality result of the patient to be examined and output the medical rationality result of the patient to be examined.

In this embodiment, data analysis can be performed on medical insurance data of a patient to be audited, and a medical action path of the patient to be audited is determined; the medical behavior path of the patient is subjected to feature extraction, and medical rationality analysis is performed according to the extracted medical behavior features of the patient to be examined so as to determine the medical rationality of the patient to be examined, so that the medical rationality can be automatically determined, and the efficiency of identifying medical fraudulent behaviors is improved.

The medical rationality determining method provided by the embodiment of the application can be deployed on any computer equipment. Optionally, the method for determining medical rationality provided by the embodiment of the application can be deployed in a cloud and used as a SaaS service. For the server-side equipment with the SaaS service, the steps in the medical rationality analysis can be executed in response to service requests of other client-side equipment. As shown in fig. 4, the method mainly includes:

401. and responding to the medical rationality analysis request, and acquiring medical insurance data of the patient to be audited.

402. And performing data analysis on the medical insurance data to determine the medical action path of the patient to be audited.

403. And performing feature extraction on the medical action path to obtain the medical action features of the patient to be audited.

404. And performing medical rationality analysis according to the medical behavior characteristics to determine the medical rationality of the patient to be examined.

405. And providing the medical rationality to the client sending out the rationality analysis request so that the client can output the medical rationality.

The medical rationality determining method provided by the embodiment can be deployed in a cloud end, and provides medical rationality analysis service for a user. The cloud end can be a private cloud of a hospital or a medical insurance system, and can also be a public cloud. For the server device deploying the medical rationality determination method, in step 401, in response to the medical rationality analysis request, medical insurance data of the patient to be examined is acquired. Optionally, the server device may provide an Application Program Interface (API) to the user, and the service requester may call the API to invoke the medical rationality analysis service. Accordingly, the medical rationality analysis request is implemented as a calling event generated by calling the API. The service requester and the server device may be communicatively connected, and the communication connection manner may refer to the related contents of the above embodiments, which is not described herein again. Alternatively, the service requester may also invoke the medical rationality analysis service through Remote Procedure Call (RPC) or Remote Direct data Access (RDMA) techniques.

For the description of steps 401 to 404, reference may be made to fig. 3 and its related contents in the alternative embodiment. Further, after the disease category of the patient to be audited is determined, the medical rationality of the patient to be audited can be provided to the client side sending the medical rationality analysis request. And for the client, receiving the medical rationality result of the patient to be examined, outputting the medical rationality of the patient to be examined, and the like. For a specific implementation of the client outputting the disease category of the patient to be audited, reference may be made to the relevant contents of the above embodiments, and details are not described here.

It should be noted that, the executing subjects of the steps of the method provided in the foregoing embodiments may be the same device, or different devices may also be used as the executing subjects of the method. For example, the execution subject of

steps

401 and 402 may be device a; for another example, the execution subject of step 401 may be device a, and the execution subject of step 402 may be device B; and so on.

In addition, in some of the flows described in the above embodiments and the drawings, a plurality of operations are included in a specific order, but it should be clearly understood that the operations may be executed out of the order presented herein or in parallel, and the sequence numbers of the operations, such as 401, 402, etc., are merely used to distinguish various operations, and the sequence numbers themselves do not represent any execution order. Additionally, the flows may include more or fewer operations, and the operations may be performed sequentially or in parallel.

Accordingly, embodiments of the present application also provide a computer-readable storage medium storing computer instructions, which, when executed by one or more processors, cause the one or more processors to perform the steps of the medical rationality method described above.

An embodiment of the present application further provides a computer program product, including: a computer program; the above-described medical rationality determining methods may be implemented when a computer program is executed by a processor. The computer program product can be medical insurance auditing system application software, web application software, saaS software and the like.

Fig. 5 is a schematic structural diagram of a computer device according to an embodiment of the present application. As shown in fig. 5, the computer apparatus includes: a memory 50a and a processor 50b; the memory 50a is used for storing computer programs.

The processor 50b is coupled to the memory 50a for executing computer programs for: acquiring medical insurance data of a patient to be audited; performing data analysis on the medical insurance data to determine a medical action path of the patient to be audited; extracting the characteristics of the medical action path to obtain the medical action characteristics of the patient to be audited; and performing medical rationality analysis according to the medical behavior characteristics to determine the medical rationality of the patient to be audited.

Optionally, when the processor 50b performs data analysis on the medical insurance data, the processor is specifically configured to: performing text analysis on the medical insurance data to determine medical entities included in the medical insurance data; determining a medical behavior corresponding to the medical entity according to the attribute of the medical entity; and sequencing the medical behaviors according to the occurrence sequence of the medical behaviors to obtain the medical behavior path of the patient to be examined.

Further, when the processor 50b performs feature extraction on the medical action path of the patient to be audited, it is specifically configured to: inputting a medical behavior path of a patient to be audited into a first network model; and in the first network model, performing feature extraction on the medical action path to obtain the medical action features of the patient to be audited.

Optionally, when the medical entity corresponding to the medical action path of the patient to be audited is input into the first network model, the processor 50b is specifically configured to: and sequentially inputting the medical entity information corresponding to the medical action path into the plurality of sub-models of the first network model according to the occurrence sequence of the medical actions.

Accordingly, when the processor 50b performs feature extraction on the medical entity corresponding to the medical action path, the feature extraction is specifically configured to: aiming at two adjacent submodels, performing feature extraction on the medical behavior feature output by the former submodel and the medical entity information input by the latter submodel by utilizing the latter submodel to obtain the medical behavior feature output by the latter submodel; and determining the medical behavior characteristics output by the plurality of sub-models as the medical behavior characteristics of the patient to be audited.

Optionally, the processor 50b is further configured to: before a medical entity corresponding to a medical action path of a patient to be audited is input into the first network model, acquiring a disease type of the patient to be audited from medical insurance data; and acquiring a feature extraction model corresponding to the disease type of the patient to be examined from the feature extraction model as a first network model.

In other embodiments, the processor 50b, when performing the medical rationality analysis based on the medical behavioral characteristics, is specifically configured to: inputting the medical behavior characteristics into a second network model; in the second network model, calculating the medical rationality probability of the patient to be examined according to the medical behavior characteristics; and determining the medical rationality of the patient to be audited according to the medical rationality probability of the patient to be audited.

Optionally, the processor 50b is further configured to: if the medical rationality of the patient to be examined is unreasonable, inputting the medical behavior characteristics into a third network model; in the third network model, performing medical rationality analysis according to the medical behavior characteristics to obtain the medical rationality of the patient to be examined; the training sample of the third network model is medical behavior characteristics obtained according to medical insurance data of similar disease types and medical insurance data of combined treatment of multiple complications; the training samples of the second network model are medical behavior features obtained according to medical insurance data of a single disease category.

In some embodiments of the present application, the processor 50b is further configured to: determining a reasonable medical action path according to the medical insurance data with reasonable known medical actions; adding unreasonable medical behaviors in the reasonable medical behavior path to obtain an unreasonable medical behavior path; and performing model training by taking the reasonable medical behavior path as a positive sample and taking the unreasonable medical behavior path as a negative sample to obtain a first network model.

In some embodiments, the processor 50b, when obtaining medical record data of a patient to be reviewed, is specifically configured to: and responding to the triggering operation of the medical record providing control, and acquiring data associated with the triggering operation as medical record data of the patient to be audited.

Alternatively, the computer device may be implemented as a terminal device on the user side. The processor 50b is further configured to: the medical rationality of the patient to be examined is displayed on the screen 50 c.

In other embodiments, the computer device may be implemented as a server-side device. The processor 50b is specifically configured to, when acquiring medical record data of a patient to be audited: analyzing the identification information of the patient to be examined from the received medical rationality analysis request; and matching the identification information of the patient to be audited in the corresponding relation between the pre-stored patient identification information and the medical insurance data to obtain the medical insurance data of the patient to be audited.

Accordingly, the processor 50b is further configured to: the medical rationality of the patient to be audited is sent to the client providing the request for initiating the medical rationality analysis through the communication component 50d for the client to output the medical rationality of the patient to be audited.

In some optional embodiments, as shown in fig. 5, the computer device may further include: power components 50e, audio components 50f, and the like. Only some of the components are shown schematically in fig. 5, and it is not meant that the computer device must contain all of the components shown in fig. 5, nor that the computer device can only contain the components shown in fig. 5.

The computer equipment provided by the embodiment of the application can perform data analysis on medical insurance data of a patient to be audited and determine a medical action path of the patient to be audited; the medical behavior path of the patient is subjected to feature extraction, and medical rationality analysis is performed according to the extracted medical behavior features of the patient to be examined so as to determine the medical rationality of the patient to be examined, so that the medical rationality can be automatically determined, and the efficiency of identifying medical fraudulent behaviors is improved.

In embodiments of the present application, the memory is used to store computer programs and may be configured to store various other data to support operations on the device on which it resides. Wherein the processor may execute a computer program stored in the memory to implement the corresponding control logic. The memory may be implemented by any type or combination of volatile or non-volatile memory devices, such as Static Random Access Memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic or optical disks.

In the embodiments of the present application, the processor may be any hardware processing device that can execute the above described method logic. Alternatively, the processor may be a Central Processing Unit (CPU), a Graphics Processing Unit (GPU), or a Micro Controller Unit (MCU); programmable devices such as Field-Programmable Gate arrays (FPGAs), programmable Array Logic devices (PALs), general Array Logic devices (GAL), complex Programmable Logic Devices (CPLDs), etc.; or Advanced Reduced Instruction Set (RISC) processors (ARM), or System On Chips (SOC), etc., but is not limited thereto.

In embodiments of the present application, the communication component is configured to facilitate wired or wireless communication between the device in which it is located and other devices. The device where the communication component is located can access a wireless network based on a communication standard, such as WiFi,2G or 3G,4G,5G or a combination thereof. In an exemplary embodiment, the communication component receives a broadcast signal or broadcast related information from an external broadcast management system via a broadcast channel. In an exemplary embodiment, the communication component may also be implemented based on Near Field Communication (NFC) technology, radio Frequency Identification (RFID) technology, infrared data association (IrDA) technology, ultra Wideband (UWB) technology, bluetooth (BT) technology, or other technologies.

In the embodiment of the present application, the display assembly may include a Liquid Crystal Display (LCD) and a Touch Panel (TP). If the display assembly includes a touch panel, the display assembly may be implemented as a touch screen to receive input signals from a user. The touch panel includes one or more touch sensors to sense touch, slide, and gestures on the touch panel. The touch sensor may not only sense the boundary of a touch or slide action, but also detect the duration and pressure associated with the touch or slide operation.

In embodiments of the present application, the power supply component is configured to provide power to the various components of the device in which it is located. The power components may include a power management system, one or more power supplies, and other components associated with generating, managing, and distributing power for the device in which the power component is located.

In embodiments of the present application, the audio component may be configured to output and/or input audio signals. For example, the audio component includes a Microphone (MIC) configured to receive an external audio signal when the device in which the audio component is located is in an operational mode, such as a call mode, a recording mode, and a voice recognition mode. The received audio signal may further be stored in a memory or transmitted via a communication component. In some embodiments, the audio assembly further comprises a speaker for outputting audio signals. For example, for devices with language interaction functionality, voice interaction with a user may be enabled through an audio component, and so forth.

It should be noted that, the descriptions of "first", "second", etc. in this document are used for distinguishing different messages, devices, modules, etc., and do not represent a sequential order, nor do they limit the types of "first" and "second".

As will be appreciated by one skilled in the art, embodiments of the present application may be provided as a method, system, or computer program product. Accordingly, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present application may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.

The present application is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the application. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.

These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.

These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.

In a typical configuration, a computing device includes one or more processors (CPUs), input/output interfaces, network interfaces, and memory.

The memory may include forms of volatile memory in a computer readable medium, random Access Memory (RAM) and/or non-volatile memory, such as Read Only Memory (ROM) or flash memory (flash RAM). Memory is an example of a computer-readable medium.

Computer-readable media, including both permanent and non-permanent, removable and non-removable media, may implement the information storage by any method or technology. The information may be computer readable instructions, data structures, modules of a program, or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static Random Access Memory (SRAM), dynamic Random Access Memory (DRAM), other types of Random Access Memory (RAM), read Only Memory (ROM), electrically Erasable Programmable Read Only Memory (EEPROM), flash memory or other memory technology, compact disc read only memory (CD-ROM), digital Versatile Discs (DVD) or other optical storage, magnetic cassettes, magnetic disk storage or other magnetic storage devices, or any other non-transmission medium that can be used to store information that can be accessed by a computing device. As defined herein, a computer readable medium does not include a transitory computer readable medium such as a modulated data signal and a carrier wave.

It should also be noted that the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrases "comprising a," "8230," "8230," or "comprising" does not exclude the presence of other like elements in a process, method, article, or apparatus comprising the element.

The above description is only an example of the present application and is not intended to limit the present application. Various modifications and changes may occur to those skilled in the art. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included in the scope of the claims of the present application.

Claims

1. A medical rationality determining method, characterized by comprising:

acquiring medical insurance data of a patient to be audited;

2. The method of claim 1, wherein the performing data analysis on the medical insurance data to determine the medical action path of the patient to be audited comprises:

performing text analysis on the medical insurance data to determine medical entities included in the medical insurance data;

determining a medical behavior corresponding to the medical entity according to the attribute of the medical entity;

and sequencing the medical behaviors according to the occurrence sequence of the medical behaviors to obtain the medical behavior path of the patient to be audited.

3. The method according to claim 1, wherein the performing feature extraction on the medical behavior path of the patient to be audited to obtain the medical behavior feature of the patient to be audited comprises:

inputting the medical action path of the patient to be audited into a first network model;

and in the first network model, performing feature extraction on the medical behavior path to obtain the medical behavior features of the patient to be audited.

4. The method according to claim 3, wherein the inputting the medical entity corresponding to the medical action path of the patient to be audited into the first network model comprises:

sequentially inputting medical entity information corresponding to the medical action path into a plurality of sub-models of the first network model according to the occurrence sequence of the medical actions;

in the first network model, performing feature extraction on the medical entity corresponding to the medical behavior path to obtain the medical behavior feature of the patient to be audited, including:

aiming at two adjacent submodels, utilizing a next submodel to perform feature extraction on medical behavior features output by a previous submodel and medical entity information input into the next submodel so as to obtain medical behavior features output by the next submodel;

and determining the medical behavior characteristics output by the plurality of sub-models as the medical behavior characteristics of the patient to be audited.

5. The method according to claim 3, wherein before inputting the medical entity corresponding to the medical action path of the patient to be audited into the first network model, the method further comprises:

acquiring the disease species of the patient to be audited from the medical insurance data;

and acquiring a feature extraction model corresponding to the disease type of the patient to be examined from the feature extraction model as the first network model.

6. The method according to any one of claims 1-5, wherein said performing a medical rationality analysis based on said medical behavioral characteristics comprises:

inputting the medical behavior feature into a second network model;

in the second network model, calculating the medical rationality probability of the patient to be audited according to the medical behavior characteristics;

and determining the medical rationality of the patient to be examined according to the medical rationality probability of the patient to be examined.

7. The method of claim 6, further comprising:

if the medical rationality of the patient to be audited is unreasonable, inputting the medical behavior characteristics into a third network model;

in the third network model, performing medical rationality analysis according to the medical behavior characteristics to obtain the medical rationality of the patient to be examined;

the training sample of the third network model is medical behavior characteristics obtained according to medical insurance data of similar disease types and medical insurance data of combined treatment of multiple complications;

the training sample of the second network model is medical behavior characteristics obtained according to medical insurance data of a single disease category.

8. The method of claim 3, further comprising:

determining a reasonable medical action path according to the known medical insurance data with reasonable medical actions;

adding unreasonable medical behaviors in the reasonable medical behavior path to obtain an unreasonable medical behavior path;

and performing model training by taking the reasonable medical behavior path as a positive sample and taking the unreasonable medical behavior path as a negative sample to obtain the first network model.

9. The method of claim 1, further comprising:

displaying the medical rationality of the patient to be audited on a screen;

alternatively, the first and second electrodes may be,

and sending the medical rationality of the patient to be examined to a client initiating a rationality analysis request so that the client can output the medical rationality.

10. A medical rationality determining method, characterized by comprising:

11. A computer device, comprising: a memory and a processor; wherein the memory is to store a computer program;

the processor is coupled to the memory for executing the computer program for performing the steps of the method of any of claims 1-10.

12. A computer program product, comprising a computer program; the computer program when executed by one or more processors may implement the method of any one of claims 1-10.