WO2020108111A1 - 医保欺诈行为的识别方法、装置、设备及可读存储介质 - Google Patents

医保欺诈行为的识别方法、装置、设备及可读存储介质 Download PDF

Info

Publication number
WO2020108111A1
WO2020108111A1 PCT/CN2019/110526 CN2019110526W WO2020108111A1 WO 2020108111 A1 WO2020108111 A1 WO 2020108111A1 CN 2019110526 W CN2019110526 W CN 2019110526W WO 2020108111 A1 WO2020108111 A1 WO 2020108111A1
Authority
WO
WIPO (PCT)
Prior art keywords
target
patient
visits
medical insurance
relationship network
Prior art date
Application number
PCT/CN2019/110526
Other languages
English (en)
French (fr)
Inventor
黄越
陈明东
Original Assignee
平安医疗健康管理股份有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 平安医疗健康管理股份有限公司 filed Critical 平安医疗健康管理股份有限公司
Publication of WO2020108111A1 publication Critical patent/WO2020108111A1/zh

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q40/00—Finance; Insurance; Tax strategies; Processing of corporate or income taxes
    • G06Q40/08—Insurance

Definitions

  • This application mainly relates to the technical field of medical systems, and in particular, to a method, device, device, and readable storage medium for identifying medical insurance fraud.
  • the main purpose of the present application is to provide a method, device, equipment and readable storage medium for identifying medical insurance frauds, aiming to solve the problem of lack of an effective identification mechanism for group medical insurance frauds in the prior art.
  • the present application provides a method for identifying medical insurance frauds.
  • the method for identifying medical insurance frauds includes the following steps:
  • the target patient pairs for which the patients focus on the visit within a preset interval determine the target patient pairs for which the patients focus on the visit within a preset interval, and count the number of visits made by the target patient pairs for the visit;
  • the visit behavior of each patient in the relationship network is identified as medical insurance fraud.
  • the present application also proposes a medical insurance fraud identification device, the medical insurance fraud identification device includes:
  • the reading module is used to read the encrypted medication information and treatment time of each patient in the preset time interval from the medical institution, and decrypt the medication information based on the decryption rules corresponding to the encryption process.
  • the decrypted processed medication information is screened, the same medication information in a preset time interval is filtered out, encrypted and stored, and the patients with the same medication information form a patient set;
  • the statistics module is used to determine the target patient pairs for which the patients focus on the consultation within a preset interval according to the time of the visit, and count the number of visits made by each target patient pair for the consultation;
  • a forming module configured to form a relationship network for each of the target patient pairs according to the number of visits, and determine whether the number of visits corresponding to the relationship network is greater than a preset value
  • the identification module is configured to identify the medical behavior of each patient in the relationship network as medical insurance fraud if the number of medical visits corresponding to the relationship network is greater than a preset value.
  • the present application also proposes a medical insurance fraud identification device, the medical insurance fraud identification device includes: a memory, a processor, a communication bus, and medical insurance fraud identification stored in the memory program;
  • the communication bus is used to implement connection communication between the processor and the memory
  • the processor is used to execute the procedure for identifying medical insurance frauds and implement the steps of the above medical insurance frauds identification method.
  • the present application also provides a readable storage medium that stores one or more programs, and the one or more programs may be executed by one or more processors to Steps for implementing the above medical insurance fraud behavior identification method.
  • FIG. 1 is a schematic flowchart of a first embodiment of a method for identifying medical insurance fraud in this application
  • FIG. 2 is a schematic diagram of functional modules of the first embodiment of the apparatus for identifying medical insurance fraud in this application;
  • FIG. 3 is a schematic diagram of the device structure of the hardware operating environment involved in the method of the embodiment of the present application.
  • This application provides a method of identifying medical insurance fraud.
  • FIG. 1 is a schematic flowchart of a first embodiment of a method for identifying medical insurance fraud based on this application.
  • the method for identifying medical insurance fraud includes:
  • Step S10 Read the encrypted medication information and the consultation time of each patient in the preset time interval from the medical institution, and decrypt each medication information based on the decryption rules corresponding to the encryption process. Screening each of the medication information, filtering out the same medication information within a preset time interval for encryption and storage, and forming a patient set of patients with the same medication information;
  • the method for identifying medical insurance fraud in this application is applied to a server, and is suitable for identifying the group's medical insurance fraud through the server; the group is composed of at least two people, and its medical insurance fraud is frequently at close intervals, and the same purchase
  • medical insurance fraud against a group is also called medical insurance cross-swapping.
  • the user corresponding to the medical insurance account used by each person to go to the medical institution to purchase medicines is treated as a patient.
  • Medical institutions include but not limited to general hospitals, traditional Chinese medicine hospitals, specialty hospitals and other types of hospitals, as well as clinics, health centers, pharmacies, etc.
  • a communication connection is established between the server and the medical institution, and a preset time interval is set in advance; the server sends a request to the medical institution to request to obtain the medication information and time of each patient in the preset time interval.
  • the medication information and consultation time of the same patient in the preset time interval are read as different medication information and consultation time, such as the preset time interval is within one week, and the patient B uses the medical insurance to buy on Monday and Wednesday If the drugs W2 and W3 are used, this will be read as the two medication information and the consultation time within the preset time interval.
  • the medication information involves the privacy of the visiting patient, in order to avoid leakage, the medication information is encrypted and stored in the medical institution; after obtaining the encrypted medication information from the medical institution, according to the encryption rules corresponding to the encryption process, Decrypt the acquired medication information.
  • the drug information that has been decrypted is filtered to filter out the same drug information in the preset time interval. After that, in order to ensure the safety of processing the same medication information, the screened same medication information is stored in a re-encrypted manner, and the visiting patients with the same medication information form a patient set. Since multiple types of the same medication information are involved in the preset time interval, there are multiple patient sets formed, and it is necessary to identify medical insurance frauds on the formed multiple patient sets.
  • Step S20 according to the time of the visit, determine the target patient pairs for which the patients focus on the visit within a preset interval, and count the number of visits made by each target patient pair for the visit;
  • the preset interval time is set in advance.
  • the difference between the consultation time of each patient is within the preset interval time, it means that each patient has visited the hospital in the near time; if the preset interval is set to one hour, the patient C1 The consultation time is 10 am, and the consultation time of patient C2 is 10:40 am on the same day; because the time interval between the two is within one hour, it is determined that the two have been seen at a near time.
  • the steps of determining the target patient pairs for which the patients are concentrated to perform the visit within a preset interval include:
  • Step S21 Make a difference between the visit times to generate a time difference between the patients in the patient set, and compare the time difference with the preset interval time to determine that each of the time differences is less than the Set the target time difference of interval time;
  • step S22 each of the visiting patients corresponding to the target time difference of the patients is determined as the target visiting patient pair for the visiting within the preset interval.
  • the difference in the time of the patients in the patient collection is generated to generate the time difference between the patients in the patient collection.
  • the patient concentration involves multiple patients, when making a difference, a patient pair can be formed between two different patients, and the difference in the time of each patient in the patient pair. If the patients involve D1, D2, and D3, they can form a patient pair between D1 and D2, a patient pair between D1 and D3, and a patient pair between D2 and D3; then D1 Make a difference between D2 and D3, D1 and D3, and D2 and D3, and generate each time difference.
  • the obtained time difference is compared with the preset interval time to obtain the magnitude relationship between each time difference and the preset interval time, and the target time difference less than the preset interval time is selected from each time difference.
  • the target time difference characterizes the patients who are concentrated in the near time for treatment.
  • the target time difference is generated by the time of the visit, and the time of the visit comes from the patient; the two patients who are concentrated in the patient and the time difference of the visit are determined to be at the preset interval
  • the patient concentration involves multiple pairs of target patients, and each target patient pair involves two different patients.
  • each patient's visit time is random, and multiple visits have been made, so that the patient may form a target patient pair multiple times within the preset time interval.
  • the preset time interval is one week, and the preset interval is one hour; where D1 and D2 are treated on Monday, Tuesday and Thursday within a week, and have Visiting time; and the time difference between Monday and Thursday is within one hour, and the time difference between Tuesday exceeds one hour; thus D1 and D2 are formed on Monday and Thursday within the preset time interval
  • the target patient is right. Count the number of times that the target patient pair is formed among the visiting patients as the number of times the target patient pair visits the clinic to determine the number of visits made by each patient who forms the target patient pair in the near time.
  • step S30 according to the number of visits, each target patient pair is formed into a relationship network, and it is determined whether the number of visits corresponding to the relationship network is greater than a preset value;
  • the patient between the target patient pairs may form other target patient pairs with other patients; as the above patient D1, D2 and D3, D1 forms the target patient pair with D2,
  • a target patient pair is also formed with D3; that is, the target patient may have an association relationship between the target patient pairs, and the association relationship is determined according to the number of visits.
  • the relationship network includes a number of patients who are in the near time, and the number of times each patient visits in the near time is the number of visits that the target patient has; that is, the formed relationship network is aligned with each target patient The number of visits is the same.
  • the number of visits reflects the number of visits made by the patient in the near time. When the number of visits is more, it means that the patient buys the same drug in the near time more frequently, the greater the possibility of being a member of the group, and there may be medical insurance fraud.
  • a preset value is set in advance; after the relationship network is formed, the number of visits corresponding to the relationship network is formed, that is, the pair of target patients who form the relationship network. The number of times is compared with the preset value to determine whether the number of visits to the relationship network is greater than the preset value.
  • Step S40 if the number of visits corresponding to the relationship network is greater than a preset value, then the visit behavior of each patient in the relationship network is identified as medical insurance fraud.
  • the relationship network when it is determined that the number of visits corresponding to the relationship network is greater than the preset value, it means that each patient in the relationship network frequently purchases the same medicine within a preset interval representing the approaching time; and the relationship network
  • the visit behavior of each patient is identified as medical insurance fraud; when it is determined that the number of visits corresponding to the relationship network is not greater than the preset value, it means that each patient in the relationship network does not exist in the near time, frequent purchase of the same drug Situation, without recognizing the medical treatment behavior of the patients referred to as medical insurance fraud.
  • the patient concentration is determined according to the consultation time Pairs of target patient visits within the preset interval, and count the number of visits made by each target patient visit; then, according to the number of visits, the target patient pairs will form a relationship network, and the number of visits corresponding to the relationship network When it is greater than the preset value, the medical behavior of each target patient in the relationship network is identified as medical insurance fraud.
  • the patients in this scheme have the same medication information for each patient, and the relationship network formed by the target patient pairs involves multiple patients who visit at close intervals; that is, the patients in the network It has the feature of buying the same medicine at the close interval.
  • the number of visits corresponding to the relationship network is greater than the preset value, it means that the patients in the relationship network frequently buy the same medicine at the close interval, so that the relationship network is judged.
  • the visiting patients have abnormal behaviors of using medical insurance, and they are recognized as medical insurance fraud.
  • the formed relationship network accurately characterizes the correlation between the consultation time and the medication information of each patient, making the group medical insurance fraud identified by the number of visits of the relationship network more accurate and effective.
  • the step of forming a relationship network between each of the target patients according to the number of visits includes:
  • Step S31 split the target patient pairs into target target patients, and allocate the number of visits corresponding to the target patient pairs among the split target patients to form target target patients Library
  • each target patient pair is split, and each target patient pair Split the patient into two target visits.
  • the number of visits corresponding to the target patient pair before splitting is allocated, so that the number of visits to the target target patient pair from each split target patient is allocated to represent the split
  • the split target patients form the target patient library, and each target patient number is used to characterize the relationship between the target patients in the target patient library.
  • Step S32 Read the number of visits between each target patient in the target patient library as the target number of visits, and form a relationship network between the target patients according to the target number of visits.
  • each target patient into a target patient library After forming each target patient into a target patient library, the number of visits between each target patient in the target patient library is read, and the read number of visits is used as the target number of visits for each target patient.
  • the relationship network between the target patients is formed; where the corresponding relationship is between the different target patients and the same target patients, and the different target patients The same number of visits to patients with the same target visit. If the target visits between the target patients N1 and N2 are read in the target visit patient library, and the target visits between the target patients N2 and N5 are 6; then N1 and N5 are based on N2 and the same The target number of visits is 6, which can form a network of relationships among N1, N2, and N5.
  • the process of forming a network of target patients is essentially the process of classifying target patients based on the number of target visits. Specifically, according to the number of visits of each target, the steps of forming a relationship network between the patients visited by each target include:
  • Step S321 Compare the target visit times to form a sequence of visit times
  • the target visits of each target visit in the target patient database are different; in order to facilitate the classification of target visits based on the number of target visits, First, compare the number of visits to each target to obtain the relationship between the number of visits to each target. The number of visits to each target is arranged according to this size relationship, and a sequence of visits is generated.
  • step S322 according to the order of the number of target visits in the sequence of the number of visits, the target patients with the same number of target visits are merged to form a relationship network.
  • the target patients with the same target visits are merged, and the target patients merged together form a relationship network.
  • the resulting relationship network also includes multiple; in order from large to small, the target patients with the maximum number of target visits are merged first, and then the The target patients with the second largest number of target visits are merged.
  • the target patient with the target number of visits is the one with fewer visits, and the frequency of the visits does not meet the requirements of fraudulent behavior that constitutes group medical insurance. Class-targeted patients were merged.
  • a preset preset value is set in advance.
  • the target number of visits in the sequence of number of visits is read and compared with the merged preset value to determine the size relationship between the two ;
  • the target patients with the target number of visits are not merged; and when the target number of visits is greater than or equal to the merged preset value, the Target patients are merged to form a network with the target number of visits.
  • the steps of combining the target patients with the same number of target visits to form each relationship network include:
  • Step q1 from each of the target patients with the same number of target visits, randomly select two of the target patients for merging to form a target patient group;
  • Step q2 judging whether there is a target patient who is consistent with the target number of target visits of the target patient group in the merged target patient pool;
  • step q3 if there is a target patient who matches the target number of patient visits of the target patient group, the target patient group is updated, and the updated target patient group is determined as a relationship network.
  • the target patients undergoing merger in the target patient pool have the same target number of visits
  • two target patients from the target patients with the same target number of visits are randomly selected and merged to form The target patient group, which has the same number of target visits as the target patient undergoing the merge operation. Then determine whether there is a target patient in the merged target patient library that is consistent with the target number of target patients in the target patient group; that is, whether there is a target patient in the target patient library that has not been merged.
  • the target patient undergoing the merge operation has the same target number of visits as the target patient undergoing the merge operation.
  • the consistent target patient is the target patient who has not performed the merge operation; the target patient who has not performed the merge operation is used Update the target patient group, that is, add the target patient who has not been merged to the target patient group for merge operation.
  • the target patient group forms a network.
  • the target patients who form each relationship network have the same target number of visits, so that each relationship network also has the target number of visits; thus reading the target number of visits forming each relationship network, and reading the read
  • the comparison of the number of target visits with the preset value that is, the comparison of the target number of visits with each network to the preset value, to determine whether the number of target visits is greater than the preset value, in order to identify each visit in the relationship network with each target number of visits Whether the member has committed medical insurance fraud of the group.
  • the step of identifying the medical behavior of each patient in the relationship network as medical insurance fraud includes:
  • Step S50 Read the patient information of each patient in the relationship network, and report the patient information to a preset agency for the preset agency to review the medical insurance qualification of each patient in the relationship network .
  • each patient in the relationship network must carry patient information during the process of using the medical insurance, read the patient information carried by the patient during the visit, and report the read patient information to the Default organization.
  • the preset institution is an institution that examines the medical insurance qualifications of the visiting patients. A communication connection is established between the preset institution and the server. After receiving the patient information reported by the server, the patient is treated according to the patient information. Of medical insurance qualifications are reviewed to prevent fraudulent medical insurance by the group.
  • the present application provides a medical insurance fraud identification device.
  • the medical insurance fraud identification device includes:
  • the reading module 10 is used to read the encrypted medication information and the consultation time of each patient in the preset time interval from the medical institution, and decrypt the medication information based on the decryption rules corresponding to the encryption process.
  • the decrypted processed medication information is screened, the same medication information in a preset time interval is filtered out, encrypted and stored, and the patients with the same medication information form a patient set;
  • the statistics module 20 is used to determine the target patient pairs for which the patients focus on the consultation within a preset interval according to the time of the visit, and count the number of visits made by each target patient pair for the consultation;
  • the forming module 30 is configured to form a relationship network for each of the target patient pairs according to the number of visits, and determine whether the number of visits corresponding to the relationship network is greater than a preset value;
  • the identification module 40 is configured to identify the medical behavior of each patient in the relationship network as medical insurance fraud if the number of medical visits corresponding to the relationship network is greater than a preset value.
  • the reading module 10 reads the medication information and consultation time of each patient in the preset time interval, and forms the patient set with the same medication information into a patient set; then according to the consultation time ,
  • the statistics module 20 determines the target patient pairs for which the patients are concentrated in the preset interval, and counts the number of visits for each target patient pair; then according to the number of visits, the forming module 30 forms a relationship between each target patient pair Network, when the number of visits corresponding to the relationship network is greater than a preset value, the identification module 40 recognizes the visit behavior of each target patient in the relationship network as medical insurance fraud.
  • the patients in this scheme have the same medication information for each patient, and the relationship network formed by the target patient pairs involves multiple patients who visit at close intervals; that is, the patients in the network It has the feature of buying the same medicine at the close interval.
  • the number of visits corresponding to the relationship network is greater than the preset value, it means that the patients in the relationship network frequently buy the same medicine at the close interval, so that the relationship network is judged.
  • the visiting patients have abnormal behaviors of using medical insurance, and they are recognized as medical insurance fraud.
  • the formed relationship network accurately characterizes the correlation between the consultation time and the medication information of each patient, making the group medical insurance fraud identified by the number of visits of the relationship network more accurate and effective.
  • the statistical module further includes:
  • a generating unit configured to make a difference between the visit times, generate a time difference between the patients in the patient set, and compare the time difference with the preset interval time to determine that each of the time differences is less than The target time difference of the preset interval time;
  • the determining unit is configured to collectively determine each of the visiting patients corresponding to the target time difference as the target visiting patient pair for the visiting within the preset interval.
  • the forming module further includes:
  • a splitting unit for splitting each of the target patient pairs into each target patient, and distributing the number of visits corresponding to the target patient pair between the split target patients Target patient database;
  • the forming unit is configured to read the number of visits between each target patient in the target patient database as the target number of visits, and form a relationship network between the target patients according to the target number of visits.
  • the forming unit is further used to:
  • the target patients with the same number of target visits are merged to form a relationship network.
  • the forming unit is further used to:
  • the target patient group is updated, and the updated target patient group is determined as a relationship network.
  • the forming module further includes:
  • the judging unit is configured to read each of the target visits forming the relationship network, and compare the read each of the target visits with the preset value to determine whether each of the target visits is greater than the preset Set value.
  • the apparatus for identifying medical insurance fraud also includes:
  • the transmission module is used to read the patient information of each patient in the relationship network and report the patient information to a preset institution for the preset agency to provide medical insurance qualifications for each patient in the relationship network For review.
  • each virtual function module of the above medical insurance fraud identification device is stored in the memory 1005 of the medical insurance fraud identification device shown in FIG. 3, and when the processor 1001 executes the medical insurance fraud identification program, the embodiment shown in FIG. 2 is realized The function of each module in
  • FIG. 3 is a schematic structural diagram of a device in a hardware operating environment involved in a method according to an embodiment of the present application.
  • the identification device for medical insurance fraud in the embodiment of the present application may be a PC (personal computer, personal computer ), or terminal devices such as smart phones, tablet computers, e-book readers, and portable computers.
  • PC personal computer, personal computer
  • terminal devices such as smart phones, tablet computers, e-book readers, and portable computers.
  • the medical insurance fraud identification device may include: a processor 1001, such as a CPU (Central Processing Unit, central processing unit), memory 1005, communication bus 1002. Among them, the communication bus 1002 is used to implement connection communication between the processor 1001 and the memory 1005.
  • the memory 1005 may be a high-speed RAM (random access memory, random access memory), can also be 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.
  • the medical insurance fraud identification device may also include a user interface, a network interface, a camera, and RF (Radio Frequency (radio frequency) circuit, sensor, audio circuit, WiFi (Wireless Fidelity, wireless broadband) module and so on.
  • the user interface may include a display (Display), an input unit such as a keyboard (Keyboard), and the optional user interface may also include a standard wired interface and a wireless interface.
  • the network interface may optionally include a standard wired interface and a wireless interface (such as a WI-FI interface).
  • the structure of the identification device for medical insurance fraud shown in FIG. 3 does not constitute a limitation on the identification device for medical insurance fraud, and may include more or fewer parts than shown, or a combination of certain Components, or different component arrangements.
  • the memory 1005 which is a readable storage medium, may include an operating system, a network communication module, and a medical insurance fraud identification program.
  • the operating system is a program to identify and control the hardware and software resources of the medical insurance fraud, and supports the operation of the medical insurance fraud identification program and other software and/or programs.
  • the network communication module is used to implement communication between various components inside the memory 1005, and to communicate with other hardware and software in the identification device for medical insurance fraud.
  • the processor 1001 is used to execute a medical insurance fraud identification program stored in the memory 1005 to implement the steps in each embodiment of the foregoing medical insurance fraud identification method.
  • the present application provides a readable storage medium, and the readable storage medium may be a non-volatile readable storage medium.
  • the readable storage medium stores one or more programs, and the one or more programs may also be executed by one or more processors to implement the steps in the embodiments of the method for identifying medical insurance fraud.
  • 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.
  • 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 readable storage medium (such as ROM) as described above /RAM, magnetic disk, and optical disk), including several instructions to enable a terminal device (which may be a mobile phone, computer, server, or network device, etc.) to perform the methods described in the embodiments of the present application.

Landscapes

  • Business, Economics & Management (AREA)
  • Accounting & Taxation (AREA)
  • Finance (AREA)
  • Engineering & Computer Science (AREA)
  • Development Economics (AREA)
  • Economics (AREA)
  • Marketing (AREA)
  • Strategic Management (AREA)
  • Technology Law (AREA)
  • Physics & Mathematics (AREA)
  • General Business, Economics & Management (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Medical Treatment And Welfare Office Work (AREA)
  • Financial Or Insurance-Related Operations Such As Payment And Settlement (AREA)

Abstract

一种医保欺诈行为的识别方法、装置、设备及可读存储介质，所述方法包括：从医疗机构中读取预设时间区间内各就诊患者的经加密处理的用药信息和就诊时间，并基于与加密处理对应的解密规则对各用药信息进行解密处理，对经解密处理的各所述用药信息进行筛选，过滤出预设时间区间内的相同用药信息进行加密储存，并将具有相同用药信息的就诊患者形成患者集(S10)；根据所述就诊时间，确定所述患者集中在预设间隔时间内进行就诊的目标就诊患者对，并统计各所述目标就诊患者对进行就诊的就诊次数(S20)；根据就诊次数，将各所述目标就诊患者对形成关系网，并判断与所述关系网对应的所述就诊次数是否大于预设值(S30)；若与关系网对应的所述就诊次数大于预设值，则将所述关系网中各就诊患者的就诊行为识别为医保欺诈行为(S40)。本方法中基于大数据分析形成的关系网，准确表征了各就诊患者在就诊时间及用药信息之间的关联关系，使得依据关系网识别的医保欺诈行为更为准确有效。

Description

医保欺诈行为的识别方法、装置、设备及可读存储介质
本申请要求于2018年11月30日提交中国专利局、申请号为201811462221.5、发明名称为“医保欺诈行为的识别方法、装置、设备及可读存储介质”的中国专利申请的优先权，其全部内容通过引用结合在申请中。
技术领域
本申请主要涉及医疗系统技术领域，具体地说，涉及一种医保欺诈行为的识别方法、装置、设备及可读存储介质。
背景技术
随着社会保障制度的发展，具有医保并使用医保就诊的人员越来越多，各人员在各医疗机构使用医保就诊或购买药品时，医疗机构的医疗人员会针对就诊人员的病症信息开出相应的医疗处方或用药信息，以对就诊人员的疾病进行治疗。
对于目前存在的一些恶意使用医保进行欺诈的团体人员，存在以团体的形式频繁购买相似性的医保药品进行出售的欺诈行为；对此类医保欺诈行为的识别在杜绝医保恶意使用方面显得尤为重要。但是目前对于团体的医保欺诈行为缺乏有效的识别机制，使得不能准确的判定医保的恶意使用。
发明内容
本申请的主要目的是提供一种医保欺诈行为的识别方法、装置、设备及可读存储介质，旨在解决现有技术中对团体的医保欺诈行为缺乏有效识别机制的问题。
为实现上述目的，本申请提供一种医保欺诈行为的识别方法，所述医保欺诈行为的识别方法包括以下步骤：
从医疗机构中读取预设时间区间内各就诊患者的经加密处理的用药信息和就诊时间，并基于与加密处理对应的解密规则对各用药信息进行解密处理，对经解密处理的各所述用药信息进行筛选，过滤出预设时间区间内的相同用药信息进行加密储存，并将具有相同用药信息的就诊患者形成患者集；
根据所述就诊时间，确定所述患者集中在预设间隔时间内进行就诊的目标就诊患者对，并统计各所述目标就诊患者对进行就诊的就诊次数；
根据所述就诊次数，将各所述目标就诊患者对形成关系网，并判断与所述关系网对应的所述就诊次数是否大于预设值；
若与所述关系网对应的所述就诊次数大于预设值，则将所述关系网中各就诊患者的就诊行为识别为医保欺诈行为。
此外，为实现上述目的，本申请还提出一种医保欺诈行为的识别装置，所述医保欺诈行为的识别装置包括：
读取模块，用于从医疗机构中读取预设时间区间内各就诊患者的经加密处理的用药信息和就诊时间，并基于与加密处理对应的解密规则对各用药信息进行解密处理，对经解密处理的各所述用药信息进行筛选，过滤出预设时间区间内的相同用药信息进行加密储存，并将具有相同用药信息的就诊患者形成患者集；
统计模块，用于根据所述就诊时间，确定所述患者集中在预设间隔时间内进行就诊的目标就诊患者对，并统计各所述目标就诊患者对进行就诊的就诊次数；
形成模块，用于根据所述就诊次数，将各所述目标就诊患者对形成关系网，并判断与所述关系网对应的所述就诊次数是否大于预设值；
识别模块，用于若与所述关系网对应的所述就诊次数大于预设值，则将所述关系网中各就诊患者的就诊行为识别为医保欺诈行为。
此外，为实现上述目的，本申请还提出一种医保欺诈行为的识别设备，所述医保欺诈行为的识别设备包括：存储器、处理器、通信总线以及存储在所述存储器上的医保欺诈行为的识别程序；
所述通信总线用于实现处理器和存储器之间的连接通信；
所述处理器用于执行所述医保欺诈行为的识别程序，实现上述医保欺诈行为的识别方法的步骤。
此外，为实现上述目的，本申请还提供一种可读存储介质，所述可读存储介质存储有一个或者一个以上程序，所述一个或者一个以上程序可被一个或者一个以上的处理器执行以用于实现上述医保欺诈行为的识别方法的步骤。
附图说明
图1是本申请的医保欺诈行为的识别方法第一实施例的流程示意图；
图2是本申请的医保欺诈行为的识别装置第一实施例的功能模块示意图；
图3是本申请实施例方法涉及的硬件运行环境的设备结构示意图。
本申请目的的实现、功能特点及优点将结合实施例，参照附图做进一步说明。
具体实施方式
应当理解，此处所描述的具体实施例仅仅用以解释本申请，并不用于限定本申请。
本申请提供一种医保欺诈行为的识别方法。
请参照图1，图1为本申请医保欺诈行为的识别方法第一实施例的流程示意图。在本实施例中，所述医保欺诈行为的识别方法包括：
步骤S10，从医疗机构中读取预设时间区间内各就诊患者的经加密处理的用药信息和就诊时间，并基于与加密处理对应的解密规则对各用药信息进行解密处理，对经解密处理的各所述用药信息进行筛选，过滤出预设时间区间内的相同用药信息进行加密储存，并将具有相同用药信息的就诊患者形成患者集；
本申请的医保欺诈行为的识别方法应用于服务器，适用于通过服务器对团体的医保欺诈行为进行识别；其中团体至少由两名人员组成，其医保欺诈行为具有频繁在临近的间隔时间点，购买相同药品的特征；如人员A1、A2、A3频繁的在9点到10点之间使用医保去购买药品W1，在某些场景下对团体的医保欺诈行为也称为医保串刷行为。将各人员去医疗机构购买药品所使用医保账户对应的用户作为就诊患者，医疗机构则包括但不限于综合医院、中医医院、专科医院等各种类型的医院，以及诊所、卫生院、药房等。服务器和医疗机构之间建立有通信连接，并预先设定预设时间区间；服务器向医疗机构发送请求，以请求获取在该预设时间区间内各就诊患者的用药信息和就诊时间。其中对于在预设时间区间内同一就诊患者的用药信息和就诊时间作为不同的用药信息和就诊时间进行读取，如预设时间区间为一周内，而就诊患者B在周一和周三均使用医保购买了药品W2和W3，则将此作为预设时间区间内的两次用药信息和就诊时间进行读取。考虑到用药信息涉及到就诊患者的隐私，为了避免泄露，医疗机构中对于用药信息经加密处理后存储；在从医疗机构中获取经加密处理的用药信息后，依据与加密处理对应的加密规则，对获取的用药信息进行解密处理。
因团体的医保欺诈行为中具有相同药品的特性，从而对经过解密处理的用药信息进行筛选，过滤出预设时间区间中的相同用药信息。此后，为了保证对相同用药信息处理的安全性，对经筛选的相同用药信息以再次加密的方式存储，并将具有该相同用药信息的就诊患者形成患者集。因预设时间区间中涉及到多种类型的相同用药信息，使得所形成的患者集也为多个，而需要对所形成的多个患者集均进行医保欺诈行为的识别。
步骤S20，根据所述就诊时间，确定所述患者集中在预设间隔时间内进行就诊的目标就诊患者对，并统计各所述目标就诊患者对进行就诊的就诊次数；
进一步地，因团体的医保欺诈行为中具有时间临近的特性，为了识别患者集中是否存在医保欺诈行为的团体，需要筛选出患者集中就诊时间在临近时间的就诊患者。预先设定预设间隔时间，当各就诊患者的就诊时间差在该预设间隔时间内，则说明各就诊患者为在临近时间进行了就诊；如设定预设间隔时间为一小时，就诊患者C1的就诊时间为上午10点，而就诊患者C2的就诊时间为同一天的上午10点40分；因两者的时间间隔在一小时之内，则判定两者为在临近时间进行了就诊。根据患者集中各就诊患者的就诊时间，可确定患者集中在预设间隔时间内进行就诊的各就诊患者，将该在临近时间进行就诊的各就诊患者作为目标就诊患者对。具体地，根据就诊时间，确定患者集中在预设间隔时间内进行就诊的目标就诊患者对的步骤包括：
步骤S21，将各所述就诊时间做差值，生成所述患者集中各就诊患者之间的时间差，并将所述时间差和所述预设间隔时间对比，确定各所述时间差中小于所述预设间隔时间的目标时间差；
步骤S22，将所述患者集中与所述目标时间差对应的各所述就诊患者，确定为在所述预设间隔时间内进行就诊的目标就诊患者对。
进一步地，为了确定患者集中各就诊患者之间的就诊时间差是否在临近时间，即预设间隔时间内；将各就诊患者的就诊时间做差值，生成患者集中各就诊患者之间的时间差。因患者集中涉及到多名就诊患者，在做差值时，可在两个不同就诊患者之间形成患者对，而对患者对中各就诊患者的就诊时间做差值。如患者集中涉及到D1、D2、和D3三名就诊患者，可在D1和D2之间形成患者对，在D1和D3之间形成患者对，在D2和D3之间形成患者对；进而对D1和D2的就诊时间做差值，D1和D3的就诊时间做差值，D2和D3的就诊时间做差值，生成各个时间差。将得到的各个时间差分别和预设间隔时间对比，得到各个时间差和预设间隔时间之间的大小关系，而从各时间差中挑选出小于预设间隔时间的目标时间差。该目标时间差表征了患者集中在临近时间进行就诊的就诊患者，目标时间差由就诊时间生成，而就诊时间来源于就诊患者；将患者集中与就诊时间差对应的两个就诊患者，确定为在预设间隔时间内进行就诊的目标就诊患者对。患者集中涉及到多对目标就诊患者对，且各个目标就诊患者对中涉及到两个不同的就诊患者。
考虑到预设时间区间内，各就诊患者的就诊时间具有随机性，且进行了多次就诊，使得就诊患者在预设时间区间内可能多次形成目标就诊患者对。如对于上述就诊患者D1、D2和D3，预设时间区间为一星期，预设间隔时间为一小时；其中D1和D2在一星期之内的星期一、星期二和星期四均进行了就诊，而具有就诊时间；且在星期一和星期四之间的时间差在一小时之内，而在星期二之间的时间差超出一小时之内；从而D1和D2在预设时间区间内的星期一和星期四均形成了目标就诊患者对。将就诊患者之间形成目标就诊患者对的次数作为目标就诊患者对进行就诊的就诊次数进行统计，以确定形成目标就诊患者对的各就诊患者在临近时间进行就诊的次数。
步骤S30，根据所述就诊次数，将各所述目标就诊患者对形成关系网，并判断与所述关系网对应的所述就诊次数是否大于预设值；
可理解地，形成目标就诊患者对之间的就诊患者，可能与其他就诊患者形成其他的目标就诊患者对；如上述就诊患者D1、D2和D3，D1在与D2形成目标就诊患者对的同时，还与D3形成了目标就诊患者对；即各目标就诊患者对之间的就诊患者可能具有关联关系，且该关联关系依据就诊次数确定。当各目标就诊患者对之间的就诊次数相同，则判定为各目标就诊患者之间具有关联关系，而将该各目标就诊患者对形成关系网。关系网中包括多个在临近时间进行就诊的就诊患者，且各就诊患者在临近时间进行就诊的次数为目标就诊患者对所具有的就诊次数；即所形成的关系网与各目标就诊患者对之间具有相同的就诊次数。就诊次数反映了就诊患者在临近时间进行就诊的次数多少，当就诊次数越多，表征就诊患者在临近时间购买相同药品越频繁，为团体成员的可能性越大，可能存在医保欺诈行为。为了反映关系网中各就诊成员在临近时间进行就诊的次数多少，预先设置有预设值；在形成关系网之后，将关系网所对应的就诊次数，即形成关系网的目标就诊患者对的就诊次数和该预设值进行对比，判断关系网的就诊次数是否大于该预设值。
步骤S40，若与所述关系网对应的所述就诊次数大于预设值，则将所述关系网中各就诊患者的就诊行为识别为医保欺诈行为。
进一步地，当判断出关系网对应的就诊次数大于预设值时，则说明关系网中各就诊患者频繁的在表征临近时间的预设间隔时间内，购买相同的药品；而将该关系网中各就诊患者的就诊行为识别为医保欺诈行为；而当判断出关系网所对应的就诊次数不大于预设值时，则说明关系网中各就诊患者不存在在临近时间内，频繁购买相同药品的情况，而不将其中就诊患者的就诊行为识别为医保欺诈行为。
本实施例的医保欺诈行为的识别方法，通过读取预设时间区间内各个就诊患者的用药信息和就诊时间，并将具有相同用药信息的就诊患者形成患者集；再根据就诊时间，确定患者集中在预设间隔时间内进行就诊的目标就诊患者对，并统计各目标就诊患者对进行就诊的就诊次数；进而依据就诊次数，将各目标就诊患者对形成关系网，当关系网所对应的就诊次数大于预设值时，则将该关系网中各目标就诊患者的就诊行为识别为医保欺诈行为。本方案的患者集中各就诊患者具有相同的用药信息，而由其中目标就诊患者对所形成的关系网，又涉及到多个在临近的间隔时间进行就诊的就诊患者；即关系网中的就诊患者具有在临近的间隔时间，购买相同药品的特征，当关系网所对应的就诊次数大于预设值，则说明关系网中就诊患者频繁在临近的间隔时间购买相同的药品，从而判定该关系网中的就诊患者存在异常使用医保的行为，而将其识别为医保欺诈行为。因所形成的关系网准确的表征了各就诊患者在就诊时间以及用药信息之间的关联关系，使得依据关系网的就诊次数所识别的团体医保欺诈行为也更为准确有效。
进一步地，在本申请医保欺诈行为的识别方法另一实施例中，所述根据所述就诊次数，将各所述目标就诊患者对形成关系网的步骤包括：
步骤S31，将各所述目标就诊患者对拆分为各目标就诊患者，并在拆分的各所述目标就诊患者之间分配与所述目标就诊患者对所对应的就诊次数，形成目标就诊患者库；
可理解地，因形成各目标就诊患者对之间的就诊患者，可能与其他就诊患者具有关联关系；为了表征该关联关系，将各个目标就诊患者对进行拆分操作，将每个目标就诊患者对拆分为两个目标就诊患者。拆分的两个目标就诊患者之间具有与其拆分前目标就诊患者对所对应的就诊次数，从而在拆分的各目标就诊患者之间分配与其来源的目标就诊患者对的就诊次数，表征拆分的各目标就诊患者之间所具有的在临近时间进行就诊的次数关系。如目标就诊患者对[M1、M2]和[M2、M4]，其中[M1、M2]所具有的就诊次数为5，而[M2、M4]所具有的就诊次数为8；则将两个目标就诊患者对分别拆分为M1和M2，以及M2和M4，且M1和M2之间所分配的就诊次数为5，M2和M4之间所分配的就诊次数为8。在所有目标就诊患者对均拆分完成后，拆分而来的目标就诊患者形成目标就诊患者库，在目标就诊患者库中用各就诊次数表征各目标就诊患者的就诊关系。
步骤S32，读取所述目标就诊患者库中各个目标就诊患者之间的就诊次数作为目标就诊次数，并根据各所述目标就诊次数，形成各所述目标就诊患者之间的关系网。
在将各个目标就诊患者形成目标就诊患者库之后，读取目标就诊患者库中各个目标就诊患者之间的就诊次数，并将该读取的就诊次数作为各目标就诊患者的目标就诊次数。依据目标就诊患者之间所具有目标就诊次数的对应关系，形成目标就诊患者之间的关系网；其中对应关系为不同的目标就诊患者之间依据同一个目标就诊患者，且该不同的目标就诊患者与该同一目标就诊患者之间的就诊次数相同的关系。如读取到目标就诊患者库中目标就诊患者N1与N2之间的目标就诊次数为6，而目标就诊患者N2与N5之间的目标就诊次数为6；则N1与N5之间依据N2以及同样的目标就诊次数6，可形成N1、N2和N5之间的关系网。该将目标就诊患者形成关系网的过程，其实质为依据目标就诊次数，对目标就诊患者进行分类的过程。具体地，根据各目标就诊次数，形成各目标就诊患者之间的关系网的步骤包括：
步骤S321，将各所述目标就诊次数进行对比，形成就诊次数序列；
因在预设时间区间内不同目标就诊患者对的就诊次数不同，使得目标就诊患者库中各目标就诊患者之间的目标就诊次数存在差异性；为了便于依据目标就诊次数对目标就诊患者进行分类，先对各目标就诊次数进行对比，得到各目标就诊次数之间的大小关系。对各目标就诊次数按照该大小关系进行排列，生成就诊次数序列。
步骤S322，按照所述就诊次数序列中目标就诊次数从大到小的顺序，将具有相同所述目标就诊次数的各所述目标就诊患者合并形成各关系网。
进一步地，按照该就诊次数序列中目标就诊次数从大到小的顺序，对各个具有相同目标就诊次数的各目标就诊患者进行合并操作，合并到一起的各目标就诊患者即形成关系网。因就诊次数序列中涉及到多个目标就诊次数，使得所形成的关系网也包括多个；按照从大到小的顺序，先将具有目标就诊次数最大值的目标就诊患者进行合并，再将具有目标就诊次数次大值的目标就诊患者进行合并。对于就诊次数序列中数值较小的目标就诊次数，具有该目标就诊次数的目标就诊患者为就诊次数较少的就诊患者，其就诊的频繁程度达不到构成团体医保欺诈行为的要求，而不对此类目标就诊患者进行合并。为了对各目标就诊患者的就诊次数多少进行判定，预先设置有合并预设值，在合并前读取就诊次数序列中的目标就诊次数和该合并预设值对比，判断两者之间的大小关系；当目标就诊次数小于该合并预设值时，则不对具有该目标就诊次数的目标就诊患者进行合并；而当目标就诊次数大于或等于该合并预设值时，则对具有该目标就诊次数的目标就诊患者进行合并，形成具有该目标就诊次数的关系网。具体地，将具有相同目标就诊次数的各目标就诊患者合并形成各关系网的步骤包括：
步骤q1，从具有相同所述目标就诊次数的各所述目标就诊患者中，任意选取两个所述目标就诊患者进行合并，形成目标就诊患者组；
步骤q2，判断合并后的所述目标就诊患者库中，是否存在和所述目标就诊患者组的目标就诊次数一致的所述目标就诊患者；
步骤q3，若存在和所述目标就诊患者组的目标就诊次数一致的所述目标就诊患者，则对所述目标就诊患者组更新，并将更新后的所述目标就诊患者组确定为关系网。
因目标就诊患者库中进行合并的各目标就诊患者之间具有相同的目标就诊次数，从而在合并时，从具有相同的目标就诊次数的目标就诊患者中任意选取两个目标就诊患者进行合并，形成目标就诊患者组，该目标就诊患者组具有和进行合并操作的目标就诊患者一致的目标就诊次数。进而判断合并后的目标就诊患者库中，是否还存在与目标就诊患者组的目标就诊次数一致的目标就诊患者；即判断目标就诊患者库中是否还存在尚未进行合并操作的目标就诊患者，该尚未进行合并操作的目标就诊患者与已经进行合并操作的目标就诊患者具有同样的目标就诊次数。当目标就诊患者库中还存在和目标就诊患者组的目标就诊次数一致的目标就诊患者，该一致的目标就诊患者即为尚未进行合并操作的目标就诊患者；用该尚未进行合并操作的目标就诊患者对目标就诊患者组进行更新，即将尚未进行合并操作的目标就诊患者添加到目标就诊患者组中，进行合并操作。当判断出目标就诊患者库中不存在和目标就诊患者组的目标就诊次数一致的目标就诊患者，则说明对具有该目标就诊次数的目标就诊患者均进行了合并操作，将经合并和更新操作的目标就诊患者组形成关系网。继续读取目标就诊患者库中具有下一项目标就诊次数的目标就诊患者进行合并，直到目标就诊患者库中大于合并预设值的目标就诊次数对应的目标就诊患者均合并完成，形成各个关系网。
因所形成的关系网涉及到多个，需要对多个关系网均判断是否存在医保欺诈行为，从而判断与关系网对应的就诊次数是否大于预设值的步骤包括：
读取形成各所述关系网的各所述目标就诊次数，并将读取的各所述目标就诊次数和所述预设值对比，判断各所述目标就诊次数是否大于预设值。
可理解地，形成各个关系网的目标就诊患者具有相同的目标就诊次数，使得各关系网同样的具有该目标就诊次数；从而读取形成各个关系网的各目标就诊次数，并将该读取的目标就诊次数和预设值对比，即将各关系网所具有的目标就诊次数和预设值对比，判断各目标就诊次数是否大于预设值，以识别具有各个目标就诊次数的关系网中的各就诊成员是否具有团体的医保欺诈行为。
进一步地，在本申请医保欺诈行为的识别方法另一实施例中，所述将所述关系网中各就诊患者的就诊行为识别为医保欺诈行为的步骤之后包括：
步骤S50，读取所述关系网中各就诊患者的患者信息，并将所述患者信息上报到预设机构，以供所述预设机构对所述关系网中各就诊患者的医保资质进行审核。
更进一步地，在经对比判断出关系网中各就诊患者的就诊行为为医保欺诈行为后，需要对关系网中各就诊患者的医保资质进行重新审核。具体地，关系网中各就诊患者在使用医保进行就诊过程中必然携带有患者信息，对该各就诊患者在就诊过程中所携带的患者信息进行读取，并将读取的该患者信息上报到预设机构。该预设机构为对就诊患者的医保资质进行审核的机构，该预设机构与服务器之间建立有通信连接，在接收到服务器上报的患者信息后，依据该患者信息对关系网中各就诊患者的医保资质进行审核，以防止团体的医保欺诈行为。
此外，请参照图2，本申请提供一种医保欺诈行为的识别装置，在本申请医保欺诈行为的识别装置第一实施例中，所述医保欺诈行为的识别装置包括：
读取模块10，用于从医疗机构中读取预设时间区间内各就诊患者的经加密处理的用药信息和就诊时间，并基于与加密处理对应的解密规则对各用药信息进行解密处理，对经解密处理的各所述用药信息进行筛选，过滤出预设时间区间内的相同用药信息进行加密储存，并将具有相同用药信息的就诊患者形成患者集；
统计模块20，用于根据所述就诊时间，确定所述患者集中在预设间隔时间内进行就诊的目标就诊患者对，并统计各所述目标就诊患者对进行就诊的就诊次数；
形成模块30，用于根据所述就诊次数，将各所述目标就诊患者对形成关系网，并判断与所述关系网对应的所述就诊次数是否大于预设值；
识别模块40，用于若与所述关系网对应的所述就诊次数大于预设值，则将所述关系网中各就诊患者的就诊行为识别为医保欺诈行为。
本实施例的医保欺诈行为的识别装置，通过读取模块10读取预设时间区间内各个就诊患者的用药信息和就诊时间，并将具有相同用药信息的就诊患者形成患者集；再根据就诊时间，统计模块20确定患者集中在预设间隔时间内进行就诊的目标就诊患者对，并统计各目标就诊患者对进行就诊的就诊次数；进而依据就诊次数，形成模块30将各目标就诊患者对形成关系网，当关系网所对应的就诊次数大于预设值时，识别模块40则将该关系网中各目标就诊患者的就诊行为识别为医保欺诈行为。本方案的患者集中各就诊患者具有相同的用药信息，而由其中目标就诊患者对所形成的关系网，又涉及到多个在临近的间隔时间进行就诊的就诊患者；即关系网中的就诊患者具有在临近的间隔时间，购买相同药品的特征，当关系网所对应的就诊次数大于预设值，则说明关系网中就诊患者频繁在临近的间隔时间购买相同的药品，从而判定该关系网中的就诊患者存在异常使用医保的行为，而将其识别为医保欺诈行为。因所形成的关系网准确的表征了各就诊患者在就诊时间以及用药信息之间的关联关系，使得依据关系网的就诊次数所识别的团体医保欺诈行为也更为准确有效。
进一步地，在本发明医保欺诈行为的识别装置另一实施例中，所述统计模块还包括：
生成单元，用于将各所述就诊时间做差值，生成所述患者集中各就诊患者之间的时间差，并将所述时间差和所述预设间隔时间对比，确定各所述时间差中小于所述预设间隔时间的目标时间差；
确定单元，用于将所述患者集中与所述目标时间差对应的各所述就诊患者，确定为在所述预设间隔时间内进行就诊的目标就诊患者对。
进一步地，在本发明医保欺诈行为的识别装置另一实施例中，所述形成模块还包括：
拆分单元，用于将各所述目标就诊患者对拆分为各目标就诊患者，并在拆分的各所述目标就诊患者之间分配与所述目标就诊患者对所对应的就诊次数，形成目标就诊患者库；
形成单元，用于读取所述目标就诊患者库中各个目标就诊患者之间的就诊次数作为目标就诊次数，并根据各所述目标就诊次数，形成各所述目标就诊患者之间的关系网。
进一步地，在本发明医保欺诈行为的识别装置另一实施例中，所述形成单元还用于：
将各所述目标就诊次数进行对比，形成就诊次数序列；
按照所述就诊次数序列中目标就诊次数从大到小的顺序，将具有相同所述目标就诊次数的各所述目标就诊患者合并形成各关系网。
进一步地，在本发明医保欺诈行为的识别装置另一实施例中，所述形成单元还用于：
从具有相同所述目标就诊次数的各所述目标就诊患者中，任意选取两个所述目标就诊患者进行合并，形成目标就诊患者组；
判断合并后的所述目标就诊患者库中，是否存在和所述目标就诊患者组的目标就诊次数一致的所述目标就诊患者；
若存在和所述目标就诊患者组的目标就诊次数一致的所述目标就诊患者，则对所述目标就诊患者组更新，并将更新后的所述目标就诊患者组确定为关系网。
进一步地，在本发明医保欺诈行为的识别装置另一实施例中，所述形成模块还包括：
判断单元，用于读取形成各所述关系网的各所述目标就诊次数，并将读取的各所述目标就诊次数和所述预设值对比，判断各所述目标就诊次数是否大于预设值。
进一步地，在本发明医保欺诈行为的识别装置另一实施例中，所述医保欺诈行为的识别装置还包括：
传输模块，用于读取所述关系网中各就诊患者的患者信息，并将所述患者信息上报到预设机构，以供所述预设机构对所述关系网中各就诊患者的医保资质进行审核。
其中，上述医保欺诈行为的识别装置的各虚拟功能模块存储于图3所示医保欺诈行为的识别设备的存储器1005中，处理器1001执行医保欺诈行为的识别程序时，实现图2所示实施例中各个模块的功能。
需要说明的是，本领域普通技术人员可以理解实现上述实施例的全部或部分步骤可以通过硬件来完成，也可以通过程序来指令相关的硬件完成，所述的程序可以存储于一种计算机可读存储介质中，上述提到的存储介质可以是只读存储器，磁盘或光盘等。
参照图3，图3是本申请实施例方法涉及的硬件运行环境的设备结构示意图。
本申请实施例医保欺诈行为的识别设备可以是PC( personal computer，个人计算机 )，也可以是智能手机、平板电脑、电子书阅读器、便携计算机等终端设备。
如图3所示，该医保欺诈行为的识别设备可以包括：处理器1001，例如CPU（Central Processing Unit，中央处理器），存储器1005，通信总线1002。其中，通信总线1002用于实现处理器1001和存储器1005之间的连接通信。存储器1005可以是高速RAM（random access memory，随机存取存储器），也可以是稳定的存储器（non-volatile memory），例如磁盘存储器。存储器1005可选的还可以是独立于前述处理器1001的存储装置。
可选地，该医保欺诈行为的识别设备还可以包括用户接口、网络接口、摄像头、RF（Radio Frequency，射频）电路，传感器、音频电路、WiFi（Wireless Fidelity，无线宽带）模块等等。用户接口可以包括显示屏（Display）、输入单元比如键盘（Keyboard），可选用户接口还可以包括标准的有线接口、无线接口。网络接口可选的可以包括标准的有线接口、无线接口（如WI-FI接口）。
本领域技术人员可以理解，图3中示出的医保欺诈行为的识别设备结构并不构成对医保欺诈行为的识别设备的限定，可以包括比图示更多或更少的部件，或者组合某些部件，或者不同的部件布置。
如图3所示，作为一种可读存储介质的存储器1005中可以包括操作系统、网络通信模块以及医保欺诈行为的识别程序。操作系统是管理和控制医保欺诈行为的识别设备硬件和软件资源的程序，支持医保欺诈行为的识别程序以及其它软件和/或程序的运行。网络通信模块用于实现存储器1005内部各组件之间的通信，以及与医保欺诈行为的识别设备中其它硬件和软件之间通信。
在图3所示的医保欺诈行为的识别设备中，处理器1001用于执行存储器1005中存储的医保欺诈行为的识别程序，实现上述医保欺诈行为的识别方法各实施例中的步骤。
本申请提供了一种可读存储介质，所述可读存储介质可以为非易失性可读存储介质。所述可读存储介质存储有一个或者一个以上程序，所述一个或者一个以上程序还可被一个或者一个以上的处理器执行以用于实现上述医保欺诈行为的识别方法各实施例中的步骤。
还需要说明的是，在本文中，术语“包括”、“包含”或者其任何其他变体意在涵盖非排他性的包含，从而使得包括一系列要素的过程、方法、物品或者装置不仅包括那些要素，而且还包括没有明确列出的其他要素，或者是还包括为这种过程、方法、物品或者装置所固有的要素。在没有更多限制的情况下，由语句“包括一个……”限定的要素，并不排除在包括该要素的过程、方法、物品或者装置中还存在另外的相同要素。
上述本申请实施例序号仅仅为了描述，不代表实施例的优劣。
通过以上的实施方式的描述，本领域的技术人员可以清楚地了解到上述实施例方法可借助软件加必需的通用硬件平台的方式来实现，当然也可以通过硬件，但很多情况下前者是更佳的实施方式。基于这样的理解，本申请的技术方案本质上或者说对现有技术做出贡献的部分可以以软件产品的形式体现出来，该计算机软件产品存储在如上所述的一个可读存储介质（如ROM/RAM、磁碟、光盘）中，包括若干指令用以使得一台终端设备（可以是手机，计算机，服务器，或者网络设备等）执行本申请各个实施例所述的方法。
以上所述仅为本申请的优选实施例，并非因此限制本申请的专利范围，凡是在本申请的构思下，利用本申请说明书及附图内容所作的等效结构变换，或直接/间接运用在其他相关的技术领域均包括在本申请的专利保护范围内。

Claims (20)

  1. 一种医保欺诈行为的识别方法，其特征在于，所述医保欺诈行为的识别方法包括以下步骤：
    从医疗机构中读取预设时间区间内各就诊患者的经加密处理的用药信息和就诊时间，并基于与加密处理对应的解密规则对各用药信息进行解密处理，对经解密处理的各所述用药信息进行筛选，过滤出预设时间区间内的相同用药信息进行加密储存，并将具有相同用药信息的就诊患者形成患者集；
    根据所述就诊时间，确定所述患者集中在预设间隔时间内进行就诊的目标就诊患者对，并统计各所述目标就诊患者对进行就诊的就诊次数；
    根据所述就诊次数，将各所述目标就诊患者对形成关系网，并判断与所述关系网对应的所述就诊次数是否大于预设值；
    若与所述关系网对应的所述就诊次数大于预设值，则将所述关系网中各就诊患者的就诊行为识别为医保欺诈行为。
  2. 如权利要求1所述的医保欺诈行为的识别方法，其特征在于，所述根据所述就诊时间，确定所述患者集中在预设间隔时间内进行就诊的目标就诊患者对的步骤包括：
    将各所述就诊时间做差值，生成所述患者集中各就诊患者之间的时间差，并将所述时间差和所述预设间隔时间对比，确定各所述时间差中小于所述预设间隔时间的目标时间差；
    将所述患者集中与所述目标时间差对应的各所述就诊患者，确定为在所述预设间隔时间内进行就诊的目标就诊患者对。
  3. 如权利要求1所述的医保欺诈行为的识别方法，其特征在于，所述根据所述就诊次数，将各所述目标就诊患者对形成关系网的步骤包括：
    将各所述目标就诊患者对拆分为各目标就诊患者，并在拆分的各所述目标就诊患者之间分配与所述目标就诊患者对所对应的就诊次数，形成目标就诊患者库；
    读取所述目标就诊患者库中各个目标就诊患者之间的就诊次数作为目标就诊次数，并根据各所述目标就诊次数，形成各所述目标就诊患者之间的关系网。
  4. 如权利要求3所述的医保欺诈行为的识别方法，其特征在于，所述根据各所述目标就诊次数，形成各所述目标就诊患者之间的关系网的步骤包括：
    将各所述目标就诊次数进行对比，形成就诊次数序列；
    按照所述就诊次数序列中目标就诊次数从大到小的顺序，将具有相同所述目标就诊次数的各所述目标就诊患者合并形成各关系网。
  5. 如权利要求4所述的医保欺诈行为的识别方法，其特征在于，所述将具有相同所述目标就诊次数的各所述目标就诊患者合并形成各关系网的步骤包括：
    从具有相同所述目标就诊次数的各所述目标就诊患者中，任意选取两个所述目标就诊患者进行合并，形成目标就诊患者组；
    判断合并后的所述目标就诊患者库中，是否存在和所述目标就诊患者组的目标就诊次数一致的所述目标就诊患者；
    若存在和所述目标就诊患者组的目标就诊次数一致的所述目标就诊患者，则对所述目标就诊患者组更新，并将更新后的所述目标就诊患者组确定为关系网。
  6. 如权利要求4所述的医保欺诈行为的识别方法，其特征在于，所述判断与所述关系网对应的所述就诊次数是否大于预设值的步骤包括：
    读取形成各所述关系网的各所述目标就诊次数，并将读取的各所述目标就诊次数和所述预设值对比，判断各所述目标就诊次数是否大于预设值。
  7. 如权利要求1所述的医保欺诈行为的识别方法，其特征在于，所述将所述关系网中各就诊患者的就诊行为识别为医保欺诈行为的步骤之后包括：
    读取所述关系网中各就诊患者的患者信息，并将所述患者信息上报到预设机构，以供所述预设机构对所述关系网中各就诊患者的医保资质进行审核。
  8. 如权利要求2所述的医保欺诈行为的识别方法，其特征在于，所述将所述关系网中各就诊患者的就诊行为识别为医保欺诈行为的步骤之后包括：
    读取所述关系网中各就诊患者的患者信息，并将所述患者信息上报到预设机构，以供所述预设机构对所述关系网中各就诊患者的医保资质进行审核。
  9. 如权利要求3所述的医保欺诈行为的识别方法，其特征在于，所述将所述关系网中各就诊患者的就诊行为识别为医保欺诈行为的步骤之后包括：
    读取所述关系网中各就诊患者的患者信息，并将所述患者信息上报到预设机构，以供所述预设机构对所述关系网中各就诊患者的医保资质进行审核。
  10. 一种医保欺诈行为的识别装置，其特征在于，所述医保欺诈行为的识别装置包括：
    读取模块，用于从医疗机构中读取预设时间区间内各就诊患者的经加密处理的用药信息和就诊时间，并基于与加密处理对应的解密规则对各用药信息进行解密处理，对经解密处理的各所述用药信息进行筛选，过滤出预设时间区间内的相同用药信息进行加密储存，并将具有相同用药信息的就诊患者形成患者集；
    统计模块，用于根据所述就诊时间，确定所述患者集中在预设间隔时间内进行就诊的目标就诊患者对，并统计各所述目标就诊患者对进行就诊的就诊次数；
    形成模块，用于根据所述就诊次数，将各所述目标就诊患者对形成关系网，并判断与所述关系网对应的所述就诊次数是否大于预设值；
    识别模块，用于若与所述关系网对应的所述就诊次数大于预设值，则将所述关系网中各就诊患者的就诊行为识别为医保欺诈行为。
  11. 如权利要求10所述的医保欺诈行为的识别装置，其特征在于，所述统计模块还包括：
    生成单元，用于将各所述就诊时间做差值，生成所述患者集中各就诊患者之间的时间差，并将所述时间差和所述预设间隔时间对比，确定各所述时间差中小于所述预设间隔时间的目标时间差；
    确定单元，用于将所述患者集中与所述目标时间差对应的各所述就诊患者，确定为在所述预设间隔时间内进行就诊的目标就诊患者对。
  12. 如权利要求10所述的医保欺诈行为的识别装置，其特征在于，所述形成模块还包括：
    拆分单元，用于将各所述目标就诊患者对拆分为各目标就诊患者，并在拆分的各所述目标就诊患者之间分配与所述目标就诊患者对所对应的就诊次数，形成目标就诊患者库；
    形成单元，用于读取所述目标就诊患者库中各个目标就诊患者之间的就诊次数作为目标就诊次数，并根据各所述目标就诊次数，形成各所述目标就诊患者之间的关系网。
  13. 如权利要求12所述的医保欺诈行为的识别装置，其特征在于，所述形成单元还用于：
    将各所述目标就诊次数进行对比，形成就诊次数序列；
    按照所述就诊次数序列中目标就诊次数从大到小的顺序，将具有相同所述目标就诊次数的各所述目标就诊患者合并形成各关系网。
  14. 如权利要求13所述的医保欺诈行为的识别装置，其特征在于，所述形成单元还用于：
    从具有相同所述目标就诊次数的各所述目标就诊患者中，任意选取两个所述目标就诊患者进行合并，形成目标就诊患者组；
    判断合并后的所述目标就诊患者库中，是否存在和所述目标就诊患者组的目标就诊次数一致的所述目标就诊患者；
    若存在和所述目标就诊患者组的目标就诊次数一致的所述目标就诊患者，则对所述目标就诊患者组更新，并将更新后的所述目标就诊患者组确定为关系网。
  15. 如权利要求13所述的医保欺诈行为的识别装置，其特征在于，所述形成模块还包括：
    判断单元，用于读取形成各所述关系网的各所述目标就诊次数，并将读取的各所述目标就诊次数和所述预设值对比，判断各所述目标就诊次数是否大于预设值。
  16. 如权利要求10所述的医保欺诈行为的识别装置，其特征在于，所述医保欺诈行为的识别装置还包括：
    传输模块，用于读取所述关系网中各就诊患者的患者信息，并将所述患者信息上报到预设机构，以供所述预设机构对所述关系网中各就诊患者的医保资质进行审核。
  17. 如权利要求11所述的医保欺诈行为的识别装置，其特征在于，所述医保欺诈行为的识别装置还包括：
    传输模块，用于读取所述关系网中各就诊患者的患者信息，并将所述患者信息上报到预设机构，以供所述预设机构对所述关系网中各就诊患者的医保资质进行审核。
  18. 如权利要求12所述的医保欺诈行为的识别装置，其特征在于，所述医保欺诈行为的识别装置还包括：
    传输模块，用于读取所述关系网中各就诊患者的患者信息，并将所述患者信息上报到预设机构，以供所述预设机构对所述关系网中各就诊患者的医保资质进行审核。
  19. 一种医保欺诈行为的识别设备，其特征在于，所述医保欺诈行为的识别设备包括：存储器、处理器、通信总线以及存储在所述存储器上的医保欺诈行为的识别程序；
    所述通信总线用于实现处理器和存储器之间的连接通信；
    所述处理器用于执行所述医保欺诈行为的识别程序，以实现以下步骤：
    从医疗机构中读取预设时间区间内各就诊患者的经加密处理的用药信息和就诊时间，并基于与加密处理对应的解密规则对各用药信息进行解密处理，对经解密处理的各所述用药信息进行筛选，过滤出预设时间区间内的相同用药信息进行加密储存，并将具有相同用药信息的就诊患者形成患者集；
    根据所述就诊时间，确定所述患者集中在预设间隔时间内进行就诊的目标就诊患者对，并统计各所述目标就诊患者对进行就诊的就诊次数；
    根据所述就诊次数，将各所述目标就诊患者对形成关系网，并判断与所述关系网对应的所述就诊次数是否大于预设值；
    若与所述关系网对应的所述就诊次数大于预设值，则将所述关系网中各就诊患者的就诊行为识别为医保欺诈行为。
  20. 一种可读存储介质，其特征在于，所述可读存储介质上存储有医保欺诈行为的识别程序，所述医保欺诈行为的识别程序被处理器执行时，实现以下步骤：
    从医疗机构中读取预设时间区间内各就诊患者的经加密处理的用药信息和就诊时间，并基于与加密处理对应的解密规则对各用药信息进行解密处理，对经解密处理的各所述用药信息进行筛选，过滤出预设时间区间内的相同用药信息进行加密储存，并将具有相同用药信息的就诊患者形成患者集；
    根据所述就诊时间，确定所述患者集中在预设间隔时间内进行就诊的目标就诊患者对，并统计各所述目标就诊患者对进行就诊的就诊次数；
    根据所述就诊次数，将各所述目标就诊患者对形成关系网，并判断与所述关系网对应的所述就诊次数是否大于预设值；
    若与所述关系网对应的所述就诊次数大于预设值，则将所述关系网中各就诊患者的就诊行为识别为医保欺诈行为。
PCT/CN2019/110526 2018-11-30 2019-10-11 医保欺诈行为的识别方法、装置、设备及可读存储介质 WO2020108111A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201811462221.5A CN109598628B (zh) 2018-11-30 2018-11-30 医保欺诈行为的识别方法、装置、设备及可读存储介质
CN201811462221.5 2018-11-30

Publications (1)

Publication Number Publication Date
WO2020108111A1 true WO2020108111A1 (zh) 2020-06-04

Family

ID=65959445

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2019/110526 WO2020108111A1 (zh) 2018-11-30 2019-10-11 医保欺诈行为的识别方法、装置、设备及可读存储介质

Country Status (2)

Country Link
CN (1) CN109598628B (zh)
WO (1) WO2020108111A1 (zh)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113626488A (zh) * 2021-08-04 2021-11-09 挂号网（杭州）科技有限公司 数据处理方法、装置、电子设备及存储介质
CN115080997A (zh) * 2022-06-02 2022-09-20 武汉金豆医疗数据科技有限公司 医保基金的移动稽查方法、装置、计算机设备及存储介质

Families Citing this family (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109598628B (zh) * 2018-11-30 2022-09-20 平安医疗健康管理股份有限公司 医保欺诈行为的识别方法、装置、设备及可读存储介质
CN111127207B (zh) * 2019-12-28 2023-06-09 哈尔滨工业大学 一种基于区块链的药品销售欺诈行为的监管系统及其监管方法
CN111275086B (zh) * 2020-01-16 2023-05-23 上海金仕达卫宁软件科技有限公司 医保群体欺诈异常行为的检测方法、装置和电子设备
CN111427926B (zh) * 2020-03-23 2023-02-03 平安医疗健康管理股份有限公司 异常医保群组识别方法、装置、计算机设备及存储介质
CN112001805B (zh) * 2020-09-07 2023-08-22 平安科技（深圳）有限公司 基于固定时间窗的医保数据处理方法、装置、设备及介质

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20080077451A1 (en) * 2006-09-22 2008-03-27 Hartford Fire Insurance Company System for synergistic data processing
CN105159948A (zh) * 2015-08-12 2015-12-16 成都数联易康科技有限公司 一种基于多特征的医疗保险欺诈检测方法
CN109598628A (zh) * 2018-11-30 2019-04-09 平安医疗健康管理股份有限公司 医保欺诈行为的识别方法、装置、设备及可读存储介质

Family Cites Families (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102945235A (zh) * 2011-08-16 2013-02-27 句容今太科技园有限公司 面向医疗保险违规和欺诈行为的数据挖掘系统
US20160144025A1 (en) * 2014-11-25 2016-05-26 Regeneron Pharmaceuticals, Inc. Methods and formulations for treating vascular eye diseases
CN104537764B (zh) * 2014-12-31 2017-06-27 浙江大学 一种医保卡异常使用的检测方法和检测系统
CN107657536B (zh) * 2017-02-20 2018-07-31 平安科技（深圳）有限公司 社保欺诈行为的识别方法和装置
CN107133438B (zh) * 2017-03-03 2018-07-13 平安医疗健康管理股份有限公司 医疗行为监控方法及装置
CN107785058A (zh) * 2017-07-24 2018-03-09 平安科技（深圳）有限公司 反欺诈识别方法、存储介质和承载平安脑的服务器
CN107609980A (zh) * 2017-09-07 2018-01-19 平安医疗健康管理股份有限公司 医疗数据处理方法、装置、计算机设备及存储介质
CN107871285A (zh) * 2017-12-06 2018-04-03 和金在线（北京）科技有限公司 一种健康保险赔欺诈及滥用的侦测方法和系统
CN108305176A (zh) * 2018-01-05 2018-07-20 上海栈略数据技术有限公司 一种基于反馈式机器学习的最优化医疗保险智能审核系统

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20080077451A1 (en) * 2006-09-22 2008-03-27 Hartford Fire Insurance Company System for synergistic data processing
CN105159948A (zh) * 2015-08-12 2015-12-16 成都数联易康科技有限公司 一种基于多特征的医疗保险欺诈检测方法
CN109598628A (zh) * 2018-11-30 2019-04-09 平安医疗健康管理股份有限公司 医保欺诈行为的识别方法、装置、设备及可读存储介质

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113626488A (zh) * 2021-08-04 2021-11-09 挂号网（杭州）科技有限公司 数据处理方法、装置、电子设备及存储介质
CN115080997A (zh) * 2022-06-02 2022-09-20 武汉金豆医疗数据科技有限公司 医保基金的移动稽查方法、装置、计算机设备及存储介质
CN115080997B (zh) * 2022-06-02 2024-01-09 武汉金豆医疗数据科技有限公司 医保基金的移动稽查方法、装置、计算机设备及存储介质

Also Published As

Publication number Publication date
CN109598628A (zh) 2019-04-09
CN109598628B (zh) 2022-09-20

Similar Documents

Publication Publication Date Title
WO2020108111A1 (zh) 医保欺诈行为的识别方法、装置、设备及可读存储介质
WO2018192472A1 (zh) 访问认证的方法及服务器、智能终端和存储装置
CN102790761B (zh) 一种区域医疗信息系统及访问权限控制方法
WO2020119385A1 (zh) 基于数据分析的处方生成监测方法、装置、设备和介质
WO2020140666A1 (zh) 数据管理方法、装置、计算机设备和存储介质
WO2020119131A1 (zh) 用药方案异常的识别方法、装置、终端及可读存储介质
WO2020119384A1 (zh) 基于大数据分析的医保异常检测方法、装置、设备和介质
CN108877920A (zh) 诊疗数据管理方法及系统
WO2020224250A1 (zh) 智能合约的触发方法、装置、设备及存储介质
WO2020078058A1 (zh) 医疗数据异常识别方法、装置、终端及存储介质
WO2020107909A1 (zh) 异常就诊费用的判断方法、装置、设备及计算机存储介质
JPH1145304A (ja) 医療支援システム
WO2020087981A1 (zh) 风控审核模型生成方法、装置、设备及可读存储介质
WO2020119402A1 (zh) 无关用药的识别方法、装置、终端及计算机可读存储介质
WO2020147384A1 (zh) 基于区块链的安全交易方法、装置、设备及存储介质
US10622104B2 (en) System and method utilizing facial recognition with online (social) network to access casualty health information in an emergency situation
WO2020119401A1 (zh) 医师违规执业的监控方法、监控服务端及存储介质
WO2020034527A1 (zh) 用户个人信息加密授权方法、装置、设备及可读存储介质
WO2020107899A1 (zh) 医疗费用预测方法、装置、设备及计算机可读存储介质
WO2022146015A1 (ko) 약물 부작용의 개인화된 확률을 산정하는 방법 및 장치
EP4046030A1 (en) System and method for determining best practices for third parties accessing a health care network
WO2017183878A1 (ko) 의료 서비스를 제공하기 위한 방법, 시스템 및 비일시성의 컴퓨터 판독 가능 기록 매체
WO2020096262A1 (ko) 전자 장치, 그의 개인 정보 제공 방법 및 이를 기록한 컴퓨터 판독 가능 기록매체
WO2020085745A1 (ko) 의료 데이터 관리 시스템 및 그 방법
WO2019132198A1 (ko) 개인정보 보호 기반 질의 처리 서비스 제공 시스템

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 19891657

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 19891657

Country of ref document: EP

Kind code of ref document: A1