WO2020119131A1 - 用药方案异常的识别方法、装置、终端及可读存储介质 - Google Patents

用药方案异常的识别方法、装置、终端及可读存储介质 Download PDF

Info

Publication number
WO2020119131A1
WO2020119131A1 PCT/CN2019/097445 CN2019097445W WO2020119131A1 WO 2020119131 A1 WO2020119131 A1 WO 2020119131A1 CN 2019097445 W CN2019097445 W CN 2019097445W WO 2020119131 A1 WO2020119131 A1 WO 2020119131A1
Authority
WO
WIPO (PCT)
Prior art keywords
details
prescription
drug
drug purchase
belong
Prior art date
Application number
PCT/CN2019/097445
Other languages
English (en)
French (fr)
Inventor
陈明东
黄越
胥畅
Original Assignee
平安医疗健康管理股份有限公司
Priority date (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 WO2020119131A1 publication Critical patent/WO2020119131A1/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

  • the present application relates to the technical field of medical insurance, and in particular, to a method, device, device, and readable storage medium for identifying an abnormal medication plan.
  • the relevant drug consumption records can be queried from the management system of each medical institution or pharmacy according to the information of the insured person, it is not certain whether the details of the drugs purchased from the insured person are consistent with the corresponding prescription details. When it is impossible to determine whether the insured person purchases prescription drugs other than prescription details, it is also impossible to confirm whether the insured person has participated in illegal drug purchases and illegal collection of outpatient co-ordination funds.
  • the main purpose of the present application is to provide a method, device, equipment and readable storage medium for identifying an abnormal medication plan, aiming to identify the behavior of an insured person who has an abnormal medication plan.
  • the present application provides a method for identifying abnormality in a medication plan.
  • the method includes the following steps:
  • the present application also provides a device for identifying abnormality in the medication plan, including:
  • An obtaining module configured to obtain the insured person's identity information, and obtain the insured person's drug purchase details within the target time period and the prescription details corresponding to the drug purchase details according to the insured person's identity information;
  • the judging module is used to judge whether there is a drug that does not belong to the prescription details in the drug purchase details;
  • the marking module is configured to mark the insured person as having an abnormal behavior of the medication plan if there is a drug in the drug purchase details that does not belong to the prescription details.
  • the present application also provides a terminal, including a processor, a memory, and computer-readable instructions stored on the memory and executable by the processor, wherein the computer-readable instructions are When the processor executes, it implements the steps of the method for recognizing the abnormality of the medication plan as described above.
  • the present application also provides a readable storage medium that stores computer readable instructions, wherein when the computer readable instructions are executed by a processor, the implementation is as described above The steps of the identification method of the abnormal medication plan.
  • this application is conducive to strengthening the supervision of the behavior of insured persons, pharmacies, and medical institutions, maintaining the stability of the drug market, and avoiding the application of outpatient co-ordination funds.
  • FIG. 1 is a schematic diagram of a hardware structure of a terminal involved in a solution of an embodiment of this application;
  • FIG. 2 is a schematic flowchart of a first embodiment of a method for identifying an abnormality in a medication plan of the present application
  • FIG. 3 is a detailed flowchart of step 10 involved in the embodiment shown in FIG. 2;
  • FIG. 4 is a detailed flowchart of step 12 involved in the embodiment shown in FIG. 3;
  • FIG. 5 is a schematic flowchart of a second embodiment of a method for identifying an abnormality in a medication plan of the present application
  • FIG. 6 is a schematic flowchart of a third embodiment of a method for identifying an abnormality in a medication plan of the application
  • FIG. 7 is a schematic flowchart of a fourth embodiment of a method for identifying an abnormality in a medication plan of this application.
  • FIG. 8 is a schematic flowchart of a fifth embodiment of a method for identifying an abnormality in a medication plan of the present application.
  • the method for recognizing the abnormality of the medication scheme involved in the embodiments of the present application is mainly applied to a terminal, which may be a device with display and processing functions such as a PC, a portable computer, and a mobile terminal.
  • FIG. 1 is a schematic structural diagram of a terminal involved in a solution according to an embodiment of this application.
  • the terminal may include a processor 1001 (for example, a CPU), a communication bus 1002, a user interface 1003, a network interface 1004, and a memory 1005.
  • the communication bus 1002 is used to realize the connection and communication between these components;
  • the user interface 1003 may include a display (Display), an input unit such as a keyboard (Keyboard);
  • the network interface 1004 may optionally include a standard wired interface, a wireless interface (Such as WI-FI interface);
  • the memory 1005 can be a high-speed RAM memory or a stable memory (non-volatile memory), such as a disk storage, the storage 1005 may optionally be a storage device independent of the foregoing processor 1001.
  • FIG. 1 does not constitute a limitation on the device, and may include more or less components than those illustrated, or combine some components, or arrange different components.
  • the network communication module is mainly used to connect to the server and perform data communication with the server; and the processor 1001 can call the computer-readable instructions stored in the memory 1005 and execute the steps of the method for identifying the abnormality of the medication regimen.
  • the embodiments of the present application provide a method for identifying abnormality in a medication plan.
  • the method for identifying the abnormal medication plan includes the following steps:
  • Step S10 Obtain the identity information of the insured person, and obtain the drug purchase details of the insured person within the target time period and the prescription details corresponding to the drug purchase details according to the identity information of the insured person;
  • the details of the insured person's diagnosis and treatment in the embodiments of the present application can be executed by the terminal, and the terminal can perform data communication with the server, and the server is in communication connection with multiple medical institution management systems or pharmacy management systems.
  • the insured person purchases medicines in a medical institution or pharmacy
  • the insured person's drug purchase details are uploaded to the medical institution management system or pharmacy management system through the settlement interface.
  • the settlement terminal can take medicines and settle according to the prescription details provided by the insured person.
  • the terminal obtains the identity information of the insured person, and obtains the drug purchase details corresponding to the target insured person from the management system of each medical institution or pharmacy through the server according to the identity information of the insured person.
  • the terminal may also be directly connected to the management system of the medical institution or pharmacy to achieve self-supervision and management within the medical institution or pharmacy.
  • the prescription details corresponding to the drug purchase details are also obtained according to the insured person's identity information.
  • the pharmacy or pharmacy needs to proceed according to the prescription details provided by the insured person.
  • the staff of the pharmacy or pharmacy uploads the prescription details based on the insured person to the management system at the time of each settlement, and the terminal obtains the purchase details of the insured person at the same time as the basis on which the drug purchase details are based Prescription details. Or take other methods to obtain the prescription details corresponding to the drug purchase details of the insured person.
  • Step S20 judging whether there is a drug that does not belong to the prescription details in the drug purchase details
  • the insured purchases prescription drugs, he needs to provide the corresponding prescription details or the source from which the prescription details can be obtained.
  • the staff of the pharmacy or pharmacy will settle the insured according to the prescription details.
  • the insured's drug purchase behavior may have illegal operations. Therefore, it is necessary to determine whether there are drugs that do not belong to the prescription details in the drug purchase details of the insured person.
  • the terminal After obtaining the drug purchase details and prescription details of the insured, the terminal obtains the drug types in the drug purchase details and prescription details according to the preset analysis mechanism, and obtains the language that the recognition model can recognize.
  • the preset analysis mechanism can be through the data of drug purchase details and prescription details obtained from the terminal (such as the core system of the human society), using the existing NLP processing flow, through the cleaning model processing, to achieve the standardization of the field matching.
  • the main working mechanism is to use the RNN model to analyze longer and more complex text content, such as diagnosis and treatment data, medicine data, disease data, etc., for the non-standard fields, while expressing both the word itself and the semantic distance.
  • a two-way RNN model is used to encode the vectors into a sentence vector matrix.
  • Each row of this matrix can be understood as a word vector —— They are sensitive to the context of the sentence.
  • the last step is called the attention mechanism. This can compress the sentence matrix into a sentence vector for prediction. By issuing diagnosis information, medicine information, and disease type information to doctors, they are matched into corresponding standardized fields.
  • a standardized detailed table of the purchased drug details and a standardized detailed table of the prescription details are formed.
  • the standardized detailed list of drug purchase details and the standardized detailed list of prescription details are input to the abnormal identification model through the observation interface and the reference interface, respectively.
  • the abnormal identification model calculates the drugs (observed values) and prescription details in the standardized detailed list of drug purchase details through calculation
  • the drugs (reference values) in the standardized detailed list are compared one by one, and the results are output.
  • the output results include the matching between the observed values and the reference values. If all observations match one of the reference values successfully, it means that there are no drugs in the drug purchase details that are not part of the prescription. If the observations do not match all the reference values, it means that there are drugs in the drug purchase details that are not part of the prescription. drug.
  • the mechanism of the anomaly recognition model may be based on the existing deviation detection model.
  • the deviation specifically refers to anomalous instances in the classified samples, special cases that do not meet the rules, or values where the observation results are inconsistent with the model predictions and change with time, etc.
  • the basic goal of deviation detection is to find meaningful differences between observations and reference values.
  • the main deviation techniques involved are clustering, sequence anomalies, nearest neighbor method, and multidimensional data analysis.
  • the algorithm for identifying anomalies in this model is mainly an anomaly detection algorithm based on distance.
  • the main core algorithm is the index-based algorithm, that is, given a data set, the index-based algorithm uses a multi-dimensional index structure R-tree, k-d tree, etc., to find the neighbor of each object within the radius d.
  • M is the maximum number of objects in the d field of the outlier data. If the object O When M+l neighbors are found, object O is not an anomaly.
  • the worst-case complexity of this algorithm is O(k*n2), k is the number of dimensions, and n is the number of objects in the data set. When k increases, the index-based algorithm has good scalability.
  • Step S30 if there is a drug that does not belong to the prescription details in the drug purchase details, it is determined whether the drugs that do not belong to the prescription details are prescription drugs.
  • the purchase of over-the-counter medicines can be purchased without a prescription, so the prescription does not contain over-the-counter medicines.
  • the judgment result shows that there are drugs that do not belong to the prescription details in the drug purchase details, it is judged whether the drugs that do not belong to the prescription details are over-the-counter drugs.
  • step S40 if the medicine that does not belong to the prescription details is a prescription medicine, it is marked that the insured person has an abnormal behavior of the medication plan.
  • Step S60 if the medicine that does not belong to the prescription details is not a prescription medicine, it is marked that the insured person does not have any abnormal behavior of the medication plan.
  • the insured when the above-mentioned drugs that are not part of the prescriptions are non-prescription drugs, the insured’s drug purchase behavior meets the relevant regulations and there are no violations. Operation behavior. At this time, the terminal executes the end instruction and marks the insured person as normal drug purchase behavior.
  • This application obtains the drug purchase details of the insured person within the target time period, and obtains the prescription details on which the drug purchase details are based on the first preset rule, and judges whether the drug purchase details exist that do not belong to the prescription details If the drug is present, the insured person is marked to have an abnormal behavior of the medication plan.
  • This application is based on artificial intelligence technology to obtain the relevant data of the insured person from the medical institution management system, and determine the insured person by judging whether there is a drug in the insured person's drug purchase details that does not match the drugs listed in the prescription details Whether there is any abnormal behavior of the medication plan, the above technical plan is conducive to strengthening the supervision of the behavior of insured persons, pharmacies, and medical institutions, maintaining the stability of the drug market, and avoiding the application of outpatient co-ordination funds.
  • FIG. 3 is a detailed flowchart of step S10 in the first embodiment of the present application.
  • Step S10 includes steps:
  • Step S11 Obtain the insured person's drug purchase details within the target time period according to the insured person's identity information
  • the terminal can obtain the drug purchase details of the insured within a certain period of time, for example, the terminal obtains the drug purchase details of the insured on the day of April 1, or the terminal obtains the drug purchases of the insured in early April .
  • the number of drug purchase details can be one or more.
  • the terminal obtains the drugs recorded in each drug purchase detail and aggregates to form a total drug purchase detail.
  • Step S12 Obtain the prescription time interval of the prescription details on which the target drug purchase details are based on the first preset rule.
  • the first preset rule stipulates: Obtain the single purchase drug details or the total purchase drug details of the drug purchase date, and then calculate the earliest issuance date after calculating the preset time period from the date of drug purchase.
  • the date and the date of drug purchase are jointly decided.
  • the date of drug purchase can be determined by selecting the date of the earliest drug purchase, or the date of the latest drug purchase, or the date of the earliest drug purchase and the date of the latest drug purchase.
  • the average value is the date of drug purchase and is not limited here.
  • the preset time period is determined according to the prescription validity period, which can be 1 month, 2 months, etc. For example, the preset time period is 2 months.
  • Step S13 Obtain the medical record of the insured person in each hospital according to the issuing time interval;
  • Step S14 Obtain the prescription details of the insured person according to the medical record.
  • the hospital related to the medical record is marked as the target hospital.
  • the number of target hospitals may be one or more.
  • the order of obtaining the prescription details from the target hospital is determined according to the distance between the target hospital and the home address of the insured person, that is, the prescription of the nearest target hospital that is related to the insured person and is located within the time interval of issuance Details, obtain prescription details one by one according to distance, and compare and judge whether there are drugs that do not belong to multiple prescription details in the target drug purchase details.
  • Step S12 includes the following steps:
  • Step S121 Obtain the date of the earliest drug purchase behavior and the date of the latest drug purchase behavior within the target time period;
  • Step S122 Obtain the earliest issuance date according to the date of the earliest drug purchase behavior minus a preset time period, wherein the issuance time interval is from the earliest issuance date to the latest drug purchase behavior occurrence date Interval.
  • the determination of the issuance time interval is particularly important.
  • the first preset rule stipulates that the issuance time interval should be performed as follows: obtain the date of the earliest drug purchase behavior recorded in the target drug purchase details and the date of the latest drug purchase behavior, and determine the earliest according to the preset time period The earliest issuance time on the day when the drug purchase takes place, the time interval formed by the earliest issuance time and the date on which the drug purchase takes place is the prescribe time interval of the prescription details on which the target drug purchase details are based, the date on which the drug purchase takes place minus the Let the time period be the earliest issuance time of the prescription details.
  • the preset time period is determined according to the prescription validity period, which can be 1 month, 2 months, etc. For example, when the preset time period is 2 months, when the single drug purchase details are obtained, the earliest and latest date of the drug purchase details are both April 1, and the prescription details on which the target drug purchase details are based are issued The time interval is from February 1 to April 1; when multiple drug purchase details are obtained, the earliest drug purchase occurs on April 1 and the latest drug purchase occurs on April 5. The time range for issuing prescription details based on the target drug purchase details is from February 1 to April 5.
  • FIG. 5 is a schematic flowchart of a second embodiment of a method for identifying abnormality in a medication plan of the present application. Based on FIG. 3, after step S13, it further includes steps: The
  • Step S15 Determine the historical disease type of the insured person according to the medical record, and analyze the symptomatic disease type of the drug in the drug purchase details;
  • the medical record includes information about the insured person, the insured person's disease, the insured person's condition, and the treatment of the doctor.
  • the symptomatic diseases corresponding to the purchased drug details are analyzed according to the types of drugs recorded in the purchased drug details.
  • Step S16 Determine whether the historical disease type is the same as the symptomatic disease type
  • the prescription drugs in the drug purchase details are determined, and the disease type to be treated is determined according to the prescription drugs.
  • the historical and symptomatic diseases of the insured it is determined whether the historical and symptomatic diseases are the same.
  • Historical diseases and symptomatic diseases preferably correspond to the broad categories of diseases rather than the disease mechanism. If the historical disease is caused by congenital heart disease, and the symptomatic disease is rheumatic heart disease, as long as they are all classified as heart diseases, the historical disease can be considered to be the same as the symptomatic disease to reduce judgment errors. If the historical disease type is the same as the symptomatic disease type, step S14 is executed to obtain the prescription details of the insured person according to the medical record.
  • Step S17 If the historical disease type is different from the symptomatic disease type, obtain the range of auxiliary medicine corresponding to the historical disease type, and determine whether there is a drug belonging to the range of auxiliary medicine in the drug purchase details;
  • auxiliary drugs are often used to help its efficacy.
  • some supplementary medicines can be used as a supplementary medicine in a department, and can also be used as a common therapeutic medicine in a department.
  • omeprazole sodium for injection is a common therapeutic medicine in the digestive department, but it is also classified as an auxiliary medicine.
  • a class of medication Therefore, when the historical disease type is different from the symptomatic disease type, it is necessary to further determine whether the auxiliary medicine corresponding to the main drug of the historical disease type in the drug purchase details to avoid misjudgment.
  • step S40 is continued to mark the insured person as having an abnormal medication plan.
  • FIG. 6 is a schematic flowchart of steps in a third embodiment of a method for identifying abnormality in a medication plan of the present application. Based on the above embodiment, step S20 includes steps: The
  • Step S21 sequentially select one of the multiple prescription details as the target prescription detail
  • the second preset rule is preferably to determine the first target hospital according to the distance between the target medical institution and the home address of the insured person, and select the latest one from the first target hospital according to the prescription issuance time
  • the prescription details are used as the target prescription details.
  • the second preset rule can also consider the frequency of the insured person's visit to the target medical institution. The more frequently the insured person visits a medical institution, the greater the chance that the medical institution will issue the prescription details on which the insured person purchases medicine.
  • Step S22 if all the drugs in the drug purchase details belong to the target prescription details, it is determined that there are no drugs that do not belong to the prescription details in the drug purchase details;
  • step S23 if part or all of the drugs in the drug purchase details do not belong to the target prescription details, it is determined that there are drugs that do not belong to the prescription details in the drug purchase details.
  • step S30 is executed, and further identification and judgment are required Whether the target prescription drug is a prescription drug. If all the drugs in the drug purchase details belong to the target prescription details, it is determined that there are no drugs that do not belong to the prescription details in the drug purchase details, and step S60 is executed to mark that the insured person does not have a medication plan Abnormal behavior.
  • FIG. 7 is a schematic flowchart of steps in a fourth embodiment of a method for identifying abnormality in a medication plan of the present application. Based on the above embodiment, step S21 includes steps: The
  • Step S211 according to a second preset rule, select a first target prescription detail from a plurality of the prescription details, and determine whether there is a drug that does not belong to the first target prescription detail in the drug purchase detail;
  • the second preset rule is preferably to determine the first target hospital according to the distance between the target medical institution and the home address of the insured person, and select the latest one from the first target hospital according to the prescription issuance time
  • the prescription details are used as the target prescription details.
  • Step S212 if there are drugs in the drug purchase details that do not belong to the first target prescription details, then the drugs in the drug purchase details that do not belong to the first target prescription details are marked and marked as pending drug;
  • step S60 is executed to mark the insured person that there is no abnormal behavior of the medicine plan.
  • the insured may have an abnormal behavior of the medication plan, and it is necessary to further identify the drugs in the drug purchase details that do not belong to the target prescription details. And mark the drugs that do not belong to the target prescription details, and mark them as pending drugs. After obtaining the drug to be examined, the compliance of the drug to be examined is judged.
  • step S213 the remaining target prescription details are selected from the remaining prescription details, and the drug to be examined is compared with the drugs recorded in the remaining target prescription details one by one.
  • Step S22 is performed according to whether there is a drug in the drug to be examined that does not belong to the remaining target prescription details, that is, whether there is a drug that does not belong to the target prescription details in the drug purchase details, and if so, step S30 is continued to determine whether the drug Whether the medicine belonging to the prescription details is a prescription medicine, if it does not exist, step S60 is executed to mark that the insured does not have any abnormal behavior of the medication plan.
  • the remaining target prescription details one by one according to the second preset rule, and compare the drug to be reviewed with the drugs recorded in the remaining target prescription details one by one until the completion of the comparison of all the pending drugs or the completion of the completion of all the remaining target prescription details.
  • the second target prescription details are determined from the remaining prescription details, and the drug to be reviewed is compared with the drugs recorded in the second target prescription details, if all the drugs to be reviewed belong to the second target For prescription details, all drugs to be reviewed are compared, and there is no need to continue to obtain the remaining target prescription details.
  • the drug that does not belong to the second target prescription detail is marked and marked as the second drug to be examined.
  • the drug in the pending drug that does not belong to the third target prescription is marked and marked as the third pending drug. And so on, until all the prescription details are compared or the drugs on the purchase details are all compared.
  • FIG. 8 is a schematic flowchart of a fifth embodiment of a method for identifying abnormality in a medication plan of the present application. Based on the above embodiment, before step S40, the method further includes steps:
  • Step S51 if the medicine that does not belong to the prescription details is a prescription medicine, obtain the identity information of the indirect insured person who has a shared medical insurance account relationship with the insured person;
  • the shared medical insurance account refers to the situation where the insured's medical insurance account can be used to purchase medicines for the indirect insured. For example, if the insured person is the father or mother of the target patient, the relationship between the insured person and the target patient can share the medical insurance account.
  • Step S52 according to the identity information of the indirect insured person and the issuing time interval, obtain indirect prescription details corresponding to the indirect insured person in each hospital;
  • Step S53 Determine whether the medicine that does not belong to the prescription details belongs to the indirect prescription details
  • step S54 if the medicine that does not belong to the prescription details of the insured person does not belong to the indirect prescription details, it is marked that the insured person has an abnormal behavior of the medication plan, and step S40 is executed.
  • step S60 is executed.
  • a report of the medication plan abnormality is generated, and warning information is sent to the medical institution or pharmacy.
  • the warning information includes information such as the medication plan abnormal report and punishment plan.
  • the abnormal report of the medication plan includes information about the drug purchase of the insured person and information about the location of the drug purchase. In this way, after the medical insurance staff reminds the object of the abnormal medication plan through the alarm, the warning information is sent to the relevant medical institution or pharmacy by mail, announcement or letter.
  • the embodiments of the present application also provide a device for identifying abnormality in the medication plan.
  • the identification device for abnormal medication plan includes:
  • An obtaining module configured to obtain the insured person's identity information, and obtain the insured person's drug purchase details within the target time period and the prescription details corresponding to the drug purchase details according to the insured person's identity information;
  • the judging module is used to judge whether there is a drug that does not belong to the prescription details in the drug purchase details;
  • the marking module is configured to mark the insured person as having an abnormal behavior of the medication plan if there is a drug in the drug purchase details that does not belong to the prescription details.
  • each module in the device for identifying abnormality of the medication plan corresponds to the steps in the embodiment of the method for identifying abnormality of the medication plan, and its function and implementation process will not be repeated here one by one.
  • embodiments of the present application also provide a computer-readable storage medium, and the computer-readable storage medium may be a non-volatile readable storage medium.
  • embodiments of the present application also provide a readable storage medium.
  • Computer readable instructions are stored on the readable storage medium, and when the computer readable instructions are executed by the processor, the steps of the method for recognizing the abnormality of the medication regimen of any of the foregoing embodiments are implemented.

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)

Abstract

本申请提供一种用药方案异常的识别方法、装置、终端及可读存储介质。用药方案异常的识别方法包括获取参保人的身份信息，并根据参保人的身份信息获取参保人在目标时间段内的购药明细以及所述购药明细所对应的处方明细；判断购药明细中是否存在不属于处方明细的药物；若存在，判断所述不属于处方明细的药物是否为处方药；若不是，则标记参保人存在用药方案异常的行为。

Description

用药方案异常的识别方法、装置、终端及可读存储介质
本申请要求于2018年12月13日提交中国专利局、申请号为201811531071.9、发明名称为“用药方案异常的识别方法、装置、终端及可读存储介质”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
本申请涉及医疗保险技术领域，尤其涉及一种用药方案异常的识别方法、装置、设备及可读存储介质。
背景技术
在医疗领域中，存在许多欺诈行为，例如个别参保人套用假处方去药店购买与其病情完全不对症的药品。或者，患者购买与处方明细中的药品种类不同的药，容易使得部分处方药品紧缺，进而导致药品市场混乱，且该参保人非法套取门诊统筹基金，严重危害到公众利益。
虽然目前能够从各个医疗机构或药店的管理系统中根据参保人的信息查询到相关的药品消费记录，但是并不能确定从参保人购买药物的明细是否与对应的处方明细相符合。在无法确定参保人是否购买处方明细以外的处方药时，也无法确认该参保人是否参与了违规购药和非法套取门诊统筹基金的行为。
发明内容
本申请的主要目的在于提供一种用药方案异常的识别方法、装置、设备及可读存储介质，旨在实现识别参保人的购药行为中存在用药方案异常的行为。
为实现上述目的，本申请提供一种用药方案异常的识别方法，所述识别方法包括以下步骤：
获取参保人的身份信息，并根据所述参保人的身份信息获取参保人在目标时间段内的购药明细以及所述购药明细所对应的处方明细；
判断所述购药明细中是否存在不属于所述处方明细的药物；
若所述购药明细中存在不属于所述处方明细的药物，判断所述不属于所述处方明细的药物是否为处方药；
若所述不属于所述处方明细的药物是处方药，则标记所述参保人存在用药方案异常的行为。
此外，为实现上述目的，本申请还提供一种用药方案异常的识别装置，包括：
获取模块，用于获取参保人的身份信息，并根据所述参保人的身份信息获取参保人在目标时间段内的购药明细以及所述购药明细所对应的处方明细；
判断模块，用于判断所述购药明细中是否存在不属于所述处方明细的药物；
标记模块，用于若所述购药明细中存在不属于所述处方明细的药物，则将所述参保人标记为存在用药方案异常的行为。
此外，为实现上述目的，本申请还提供一种终端，包括处理器、存储器、以及存储在所述存储器上并可被所述处理器执行的计算机可读指令，其中所述计算机可读指令被所述处理器执行时，实现如上述所述的用药方案异常的识别方法的步骤。
此外，为实现上述目的，本申请还提供一种可读存储介质，所述可读存储介质上存储有计算机可读指令，其中所述计算机可读指令被处理器执行时，实现如上述所述的用药方案异常的识别方法的步骤。
本申请通过上述技术方案有利于加强对参保人、药店、医疗机构的行为进行监督，维护药品市场稳定，避免门诊统筹基金被套用。
附图说明
图1为本申请实施例方案中涉及的终端的硬件结构示意图；
图2为本申请用药方案异常的识别方法第一实施例的流程示意图；
图3为图2所示实施例涉及的步骤10的细化流程示意图；
图4为图3所示实施例涉及的步骤12的细化流程示意图；
图5为本申请用药方案异常的识别方法第二实施例的流程示意图；
图6为本申请用药方案异常的识别方法第三实施例的流程示意图；
图7为本申请用药方案异常的识别方法第四实施例的流程示意图；
图8为本申请用药方案异常的识别方法第五实施例的流程示意图。
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。
本申请实施例涉及的用药方案异常的识别方法主要应用于终端，该终端可以是PC、便携计算机、移动终端等具有显示和处理功能的设备。
参照图1，图1为本申请实施例方案中涉及的终端结构示意图。本申请实施例中，终端可以包括处理器1001（例如CPU），通信总线1002，用户接口1003，网络接口1004，存储器1005。其中，通信总线1002用于实现这些组件之间的连接通信；用户接口1003可以包括显示屏（Display）、输入单元比如键盘（Keyboard）；网络接口1004可选的可以包括标准的有线接口、无线接口（如WI-FI接口）；存储器1005可以是高速RAM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。
本领域技术人员可以理解，图1中示出的硬件结构并不构成对设备的限定，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。
继续参照图1，图1中作为一种可读存储介质的存储器1005可以包括操作系统、网络通信模块以及计算机可读指令。
在图1中，网络通信模块主要用于连接服务器，与服务器进行数据通信；而处理器1001可以调用存储器1005中存储的计算机可读指令，并执行用药方案异常的识别方法的步骤。
本申请实施例提供了一种用药方案异常的识别方法。
请参阅图2，本申请实施例中，用药方案异常的识别方法包括步骤：
步骤S10，获取参保人的身份信息，并根据所述参保人的身份信息获取参保人在目标时间段内的购药明细以及所述购药明细所对应的处方明细；
本申请实施例中的参保人诊疗明细可以通过终端执行，终端能够与服务器进行数据通信，服务器与多个医疗机构管理系统或药店管理系统之间通信连接。参保人在医疗机构或者药店购买药品时，参保人的购药明细通过结算接口端上传至医疗机构管理系统或者药店管理系统。参保人使用医保账户结算医疗账单时，结算端能够根据参保人提供的处方明细，并进行取药和结算。
终端通过获取参保人的身份信息，并根据参保人的身份信息通过服务器从各个医疗机构或药店的管理系统中获取与目标参保人对应的购药明细。在其他实施例中，终端还可以与医疗机构或者药店的管理系统进行直接连接，以实现医疗机构或药店内部自我监督管理。
获取到参保人在目标时间段内的购药明细时，同时根据参保人的身份信息获取购药明细所对应的处方明细。参保人在购买处方药时，药房或者药店需要根据参保人提供的处方明细来进行。此时，药房或者药店的工作人员在每一次结算时，同时将参保人所依据的处方明细上传至管理系统，终端获取参保人的购药明细的同时获取该次购药明细所依据的处方明细。或者采取其他方式获取与参保人的购药明细对应的处方明细。
步骤S20，判断所述购药明细中是否存在不属于所述处方明细的药物；
参保人在购药处方药时，需要提供相应的处方明细或者提供能够取得所依据的处方明细的来源，药房或药店的工作人员根据处方明细来为参保人结算。当参保人的购药明细中存在不属于其提供的处方明细的药物时，则参保人的购药行为中可能存在违规操作。因此，需要判断参保人的购药明细中是否存在不属于处方明细的药物。
终端获取到参保人的购药明细和处方明细后，按照预设分析机制获取购药明细和处方明细中药物类型，得到识别模型能够识别的语言。预设分析机制可以是通过针对终端（如人社核心系统）获取的购药明细和处方明细数据，利用现有的NLP处理流程，通过清洗模型的处理，实现对字段标准化的匹配。主要工作机制是，对不规范的字段，在既表示词本身又可以考虑语义距离的要求下，利用RNN模型分析更长更复杂的文本内容，如诊疗数据、药品数据、疾病数据等。将文本用一个向量的序列表示之后，使用双向RNN模型将向量编码为一个句子向量矩阵。这个矩阵的每一行可以理解为词向量 —— 它们对句子的上下文敏感。最后一步被称为注意力机制。这可以将句子矩阵压缩成一个句子向量，用于预测。通过对医生出具诊断信息、药品信息、病种信息匹配到相应的标准化字段中。
分析得到的购药明细中各药物的标准化字段以及处方明细中各药物的标准化字段后，形成购药明细的标准化明细表和处方明细的标准化明细表。将购药明细的标准化明细表和处方明细的标准化明细表分别通过观察接口和参照接口输入至异常识别模型，异常识别模型通过运算将购药明细的标准化明细表中药物（观察值）与处方明细的标准化明细表的药物（参照值）进行逐一对比，并输出结果，输出结果包括观察值与参照值的匹配情况。若所有观察值与其中一个参照值匹配成功，则说明购药明细内中不存在不属于处方明细的药物，若观察值与所有参照值不匹配，则说明购药明细中存在不属于处方明细的药物。
异常识别模型的机制可以是基于现有的偏差检测模型，偏差具体指分类样本中的反常实例、不满足规则的特例，或者观测结果与模型预测值不一致并随时间的变化的值等等。偏差检测的基本目标是寻找观测结果与参照值之间有意义的差别，涉及的主要的偏差技术有聚类、序列异常、最近邻居法、多维数据分析等。本模型识别异常的算法主要是基于距离的异常点检测算法。其中主要的核心算法是基于索引的算法，即给定一个数据集合，基于索引的算法采用多维索引结构 R-树，k-d树等，来查找每个对象在半径 d范围内的邻居。假设 M为异常点数据的d领域内的最大对象数目。如果对象O的 M+l个邻居被发现，则对象O就不是异常点。这个算法在最坏情况下的复杂度为O(k*n2)，k为维数，n为数据集合中对象的数目。当k增加时，基于索引的算法具有良好的扩展性。
步骤S30，若所述购药明细中存在不属于所述处方明细的药物，判断所述不属于所述处方明细的药物是否为处方药。
由于参保人在购药时，可能同时购药处方药和非处方药，非处方药的购药不需要处方明细亦能购买，因此处方明细中不含有非处方药。为了保障参保人购药非处方药的权利，当判断结果显示购药明细中存在不属于处方明细的药物时，判断不属于处方明细的药物是否为非处方药。
步骤S40，若所述不属于所述处方明细的药物是处方药，则标记所述参保人存在用药方案异常的行为。
当判断结果显示上述不属于处方明细的药物不是非处方药，即说明上述不属于处方明细的药物是处方药，则参保人在没有依据相关处方明细购买处方药的行为是不符合规定的，则标记参保人存在用药方案异常的行为，以利于终端便于识别违规行为。
步骤S60，若所述不属于所述处方明细的药物不是处方药，则标记所述参保人不存在用药方案异常的行为。
若上述不属于处方明细的药物是非处方药，由于医保并没有对非处方药的购买进行限定，因此当上述不属于处方明细的药物是非处方药时，则参保人的购药行为符合相关规定，不存在违规操作行为，此时终端执行结束指令，并将参保人标记为正常购药行为。
本申请获取参保人在目标时间段内的购药明细，并根据第一预设规则获取所述购药明细所依据的处方明细，判断所述购药明细中是否存在不属于所述处方明细的药物，若存在，则标记所述参保人存在用药方案异常的行为。本申请基于人工智能技术能够从医疗机构管理系统中获取参保人的相关数据，通过判断参保人的购药明细中是否存在与其处方明细所记载的药物不相符的药物，来判断参保人是否存在用药方案异常的行为，通过上述技术方案有利于加强对参保人、药店、医疗机构的行为进行监督，维护药品市场稳定，避免门诊统筹基金被套用。
参照图3，图3为本申请第一实施例中步骤S10的细化流程图，步骤S10包括步骤：
步骤S11，根据参保人的身份信息获取参保人在目标时间段内的购药明细；
终端可以获取某一时间段内参保人的购药明细，例如，终端获取参保人在4月1号当天发生的购药明细，或者终端获取参保人在4月上旬发生的购药明细。购药明细的数量可以为一个或者多个。当在目标时间段内存在多个购药明细时，终端获取各个购药明细中记载的药物并汇总形成总购药明细。步骤S12，根据第一预设规则获取目标购药明细所依据的处方明细的开具时间区间。
得到单个购药明细或者总购药明细后，根据第一预设规则获取目标购药明细（单个购药明细或者总购药明细）的所依据的处方明细的开具时间区间。第一预设规则规定：获取单个购药明细或者总购药明细的购药之日，然后在购药之日起向前推算预设时间段后得到最早开具之日，开具时间区间由最早开具之日和购药之日共同决定。购药之日的确定方式可以是选取最早的购药行为发生之日，或最晚的购药行为发生之日，或者最早的购药行为发生之日与最晚的购药行为发生之日的平均值作为购药之日，在此不限定。此外，预设时间段根据处方有效期限确定，可以为1个月、2个月等等。例如，预设时间段为2个月时。
步骤S13，根据所述开具时间区间获取所述参保人在各个医院的就诊记录；
步骤S14，根据所述就诊记录获取所述参保人的处方明细。
确定处方明细的开具时间区间后，从各个医院获取与参保人相关的位于开具时间区间内的就诊记录。若在开具时间区间内存在参保人的就诊记录，则标记该就诊记录相关的医院为目标医院。根据开具时间区间以及参保人的信息从目标医院的管理系统中获取参保人的处方明细。可以理解的是，所获取的参保人的处方明细的数量为一个或多个。获取到与参保人相关的位于开具时间区间内的多个处方明细后，将各个处方明细的药物进行汇总。目标医院的数量可能为一个或多个。可选地，按照目标医院距离参保人的家庭住址的距离大小确定从目标医院的获取处方明细的先后，即先获取距离最近的目标医院的与参保人相关的位于开具时间区间内的处方明细，按照距离远近逐个获取处方明细，并对比判断目标购药明细中是否存在不属于多个处方明细的药物。
参照图4，图4为上述实施例中步骤S12的细化流程图，步骤S12包括以下步骤：
步骤S121，获取在目标时间段内的最早购药行为发生之日和最晚购药行为发生之日；
步骤S122，根据所述最早购药行为发生之日减去预设时间段得到最早开具之日，其中所述开具时间区间为所述最早开具之日至最晚购药行为发生之日组成的时间区间。
为了提高识别结果的准确性，开具时间区间的确定尤为重要。可选地，第一预设规则规定开具时间区间按照如下步骤进行：获取目标购药明细中记载的最早购药行为发生之日和最晚购药行为发生之日，按照预设时间段确定最早购药行为发生之日的最早开具时间，最早开具时间与购药行为发生之日所形成的时间区间为目标购药明细所依据的处方明细的开具时间区间，购药行为发生之日减去预设时间段为处方明细的最早开具时间。预设时间段根据处方有效期限确定，可以为1个月、2个月等等。例如，预设时间段为2个月时，当得到单个购药明细时，购药明细的最早和最晚发生之日均为4月1日，则目标购药明细所依据的处方明细的开具时间区间为2月1日至4月1日；当得到多个购药明细时，最早购药行为发生之日为4月1日，最晚购药行为发生之日为4月5日，则目标购药明细所依据的处方明细的开具时间区间为2月1日至4月5日。
参照图5，图5为本申请用药方案异常的识别方法第二实施例的流程示意图。基于图3，步骤S13之后还包括步骤：
步骤S15，根据所述就诊记录确定所述参保人的历史病种，并分析所述购药明细中的药物的对症病种；
从各个医疗机构获取到与参保人相关的就诊记录之后，根据就诊记录确定参保人的历史病种。就诊记录包括有参保人信息、参保人病种、参保人病情以及处理医师等等信息。同时根据购药明细中记录的药物类型分析出购药明细所对应的对症病种。
步骤S16，判断所述历史病种与所述对症病种是否相同；
根据购药明细对症的病种前，确定购药明细中的处方药，由根据处方药确定其治疗的病种。根据获得的参保人的历史病种和对症病种，判断历史病种与对症病种是否相同。历史病种与对症病种优选对应病种大类，而不是病情机理。如历史病种是由先天性心脏病，而对症病种是风湿性心脏病，只要均归于心脏病类，即可认为历史病种与对症病种相同，以减少判断误差。若所述历史病种与所述对症病种相同，则执行步骤S14，根据所述就诊记录获取所述参保人的处方明细。
步骤S17，若所述历史病种与所述对症病种不相同，则获取所述历史病种对应的辅助用药范围，判断所述购药明细中是否存在属于所述辅助用药范围的药物；
需要注意的是，为了提高主治药物的治疗效果，往往会搭配辅助用药以助其疗效。然而，有些辅助用药除了可以作为一科室的辅助用药使用外，还可以作为某一科室的常用治疗药物使用，如注射用奥美拉唑钠为消化科常用治疗药物，但其也被归入辅助用药一类。因此，当历史病种与对症病种不相同时，需要进一步判断购药明细中是否与历史病种的主治药物对应的辅助用药，以避免出现误判的情况。若购药明细中存在属于历史病种的辅助用药范围的药物，则说明购药明细中存在治疗历史病种的药物，参保人不存在用药方案异常的行为，若购药明细中不存在属于历史病种的辅助用药范围的药物，则说明参保人可能存在用药方案异常的行为。需要注意的是，由于当参保人的历史病种与参保人的购药明细的对症病种都不相同时，且购药明细中不存在历史病种的辅助用药时，则直接可标记参保人存在用药方案异常的行为，不用再进行处方明细的查找了，有利于加快识别进度，继续执行步骤S40，标记所述参保人存在用药方案异常的行为。
参照图6，图6为本申请用药方案异常的识别方法第三实施例步骤的流程示意图。基于上述实施例，步骤S20包括步骤：
步骤S21，按照第二预设规则依次选出多份所述处方明细中的一份处方明细作为目标处方明细；
由于可能存在多张处方明细，从多个目标医疗机构获取到与参保人相关的多份处方明细后，根据第二预设规则从多份处方明细中筛选出一份处方明细作为目标处方明细。第二预设规则优选是，根据目标医疗机构距离参保人的家庭住址的距离大小确定第一目标医院，从第一目标医院中根据处方明细的开具时间选取处方明细开具时间最晚的一份处方明细作为目标处方明细。第二预设规则除了考虑目标医疗机构的距离、处方明细的开具时间之外，还可以考虑参保人在目标医疗机构的就诊频率。参保人在某医疗机构的就诊频率越大，该医疗机构开具参保人购药所依据的处方明细的机会越大。
步骤S22，若所述购药明细中的药物全部属于所述目标处方明细，则判定所述购药明细中不存在不属于所述处方明细的药物；
步骤S23，若所述购药明细中的药物部分或全部不属于所述目标处方明细，则判定所述购药明细中存在不属于所述处方明细的药物。
根据上述第二预设规则获取到目标处方明细后，判断购药明细中是否存在不属于目标处方明细的药物。若所述购药明细中的药物中存在不属于所述目标处方明细的药物，则判定所述购药明细中存在不属于所述处方明细的药物，执行步骤S30，需要进一步识别和判断不属于目标处方明细的药物是否为处方药。若所述购药明细中的药物全部属于所述目标处方明细，则判定所述购药明细中不存在不属于所述处方明细的药物，执行步骤S60，标记所述参保人不存在用药方案异常的行为。
参照图7，图7为本申请用药方案异常的识别方法第四实施例步骤的流程示意图。基于上述实施例，步骤S21包括步骤：
步骤S211，按照第二预设规则从多份所述处方明细中选出第一目标处方明细，判断所述购药明细中是否存在不属于所述第一目标处方明细的药物；；
根据第二预设规则从多份处方明细中筛选出一份处方明细作为第一目标处方明细后，判断所述购药明细中是否存在不属于所述第一目标处方明细的药物。第二预设规则优选是，根据目标医疗机构距离参保人的家庭住址的距离大小确定第一目标医院，从第一目标医院中根据处方明细的开具时间选取处方明细开具时间最晚的一份处方明细作为目标处方明细。
步骤S212，若所述购药明细中存在不属于所述第一目标处方明细的药物，则对所述购药明细中不属于所述第一目标处方明细的药物进行标记，并标记为待审药物；
若购药明细的全部药物属于第一目标处方明细，则执行步骤S60，标记参保人不存在用药方案异常的行为。
若购药明细中存在不属于第一目标处方明细的药物时，则说明参保人可能存在用药方案异常的行为，需要进一步识别购药明细中不属于目标处方明细的药物。并对不属于目标处方明细的药物进行标记，且标记为待审药物。获得待审药物之后，对待审药物的合规性进行判别。
步骤S213，从剩余所述处方明细中选出剩余目标处方明细，并将所述待审药物与剩余目标处方明细记载的药物逐张进行对比。
根据待审药物中是否存在不属于剩余目标处方明细记载的药物来执行步骤S22，即判断购药明细中是否存在不属于目标处方明细的药物，若存在，则继续执行步骤S30，判断所述不属于处方明细的药物是否为处方药，若不存在，则执行步骤S60，标记参保人不存在用药方案异常的行为。
按照第二预设规则逐张获取剩余目标处方明细，并将待审药物与剩余目标处方明细记载的药物逐张进行对比，直至待审药物全部对比完毕结束或者剩余目标处方明细全部对比完毕结束。具体地，按照第二预设规则从剩余处方明细中确定第二张目标处方明细，将待审药物与第二张目标处方明细中记载的药物进行对比，若待审药物全部属于第二张目标处方明细，则待审药物全部对比完毕，不需要继续获取剩余目标处方明细。若待审药物中存在不属于第二张目标处方明细的药物，则标记待审药物中不属于第二张目标处方明细的药物，并标记为第二待审药物。继续按照第二预设规则从未被对比的剩余目标处方明细中确定第三张目标处方明细，将第二待审药物与第三张目标处方明细中记载的药物进行对比，若第二待审药物全部属于第三张目标处方明细，则第二待审药物全部对比完毕，不需要继续获取剩余处方明细。若第二待审药物中存在不属于第三张目标处方明细的药物，则标记待审药物中不属于第三张目标处方明细的药物，并标记为第三待审药物。依次类推，直至全部处方明细均经过对比或者购药明细上的药物全部对比完毕时结束。
判断所述待审药物中是否存在不属于剩余所述目标处方明细中的药物，即若全部处方明细均经过对比之后，购药明细中记载的药物仍有被标记为待审药物时，则说明待审药物中存在不属于剩余处方明细中药物，参保人可能存在用药方案异常的行为。若待审药物与剩余处方明细对比时，待审药物全部对比完毕，则说明待审药物中不存在不属于剩余处方明细的药物，参保人不存在用药方案异常的行为。因此，根据未被对比的待审药物来判断购药明细中是否存在不属于所述目标处方明细的药物。
参照图8，图8为本申请用药方案异常的识别方法第五实施例的流程示意图。基于上述实施例，步骤S40之前还包括步骤：
步骤S51，若所述不属于所述处方明细的药物是处方药，则获取与所述参保人为共用医保账户关系的间接参保人的身份信息；
若上述不属于处方明细的药物是处方药，则获取与参保人为共用医保账户关系的潜在参保人的身份信息。共用医保账户是指能够使用参保人的医保账户为间接参保人购买药品的情况。例如参保人是目标患者的父亲或母亲，则参保人与目标患者之间为能够共用医保账户的关系。
步骤S52，根据所述间接参保人的身份信息和所述开具时间区间，获取各个医院的与所述间接参保人对应的间接处方明细；
确定好潜在参保人后，根据间接参保人的身份信息和获得的开具时间区间，从各个医院中获取与所述间接参保人对应的、且位于所述开具时间区间的间接处方明细。
步骤S53，判断所述不属于所述处方明细的药物是否属于所述间接处方明细；
步骤S54，若所述不属于参保人的所述处方明细的药物也不属于所述间接处方明细，则标记所述参保人存在用药方案异常的行为，执行步骤S40。
若所述不属于参保人的所述处方明细的药物属于所述间接处方明细，则标记所述参保人不存在用药方案异常的行为，执行步骤S60。
将不属于参保人的处方明细的药物标记出来，记为待对比药物，将待对比药物与间接处方明细中记录的药物进行对比，判断待对比药物是否属于间接处方明细，若所述待对比药物同时也不属于所述间接处方明细，则说明与参保人的处方明细不符合的药物也不属于间接参保人人所用到的药物，则标记所述参保人存在用药方案异常的行为。若所述待对比药物同时属于所述间接处方明细，则说明与参保人的处方明细不符合的药物是间接参保人人所用到的药物，则说明参保人的参保行为正常。
当参保人被标记为存在用药方案异常的行为时，生成用药方案异常报告，并向医疗机构或者药店发送警示信息，警示信息包括药方案异常报告、处罚方案等信息。用药方案异常报告包括参保人的购药信息、购药地点信息等。如此，医保工作人员通过警报提醒发现用药方案异常的对象之后，将警示信息以为邮件、公告或者书信等方式发送至相关医疗机构或者药店。
此外，本申请实施例还提供一种用药方案异常的识别装置。用药方案异常的识别装置包括：
获取模块，用于获取参保人的身份信息，并根据所述参保人的身份信息获取参保人在目标时间段内的购药明细以及所述购药明细所对应的处方明细；
判断模块，用于判断所述购药明细中是否存在不属于所述处方明细的药物；
标记模块，用于若所述购药明细中存在不属于所述处方明细的药物，则将所述参保人标记为存在用药方案异常的行为。
其中，上述用药方案异常的识别装置中各个模块的功能实现与上述用药方案异常的识别方法实施例中各步骤相对应，其功能和实现过程在此处不再一一赘述。
此外，本申请实施例还提供一种计算机可读存储介质，所述计算机可读存储介质可以为非易失性可读存储介质。
此外，本申请实施例还提供一种可读存储介质。可读存储介质上存储有计算机可读指令，其中计算机可读指令被处理器执行时，实现上述任一实施例的用药方案异常的识别方法的步骤。
其中，计算机可读指令被执行时所实现的方法可参照本申请多用药方案异常的识别方法的各个实施例，此处不再赘述。
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、物品或者系统不仅包括那些要素，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、物品或者系统所固有的要素。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、物品或者系统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目前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，空调器，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，凡是利用本申请说明书及附图内容所作的等效结构或等效流程变换，或直接或间接运用在其他相关的技术领域，均同理包括在本申请的专利保护范围内。

Claims (20)

  1. 一种用药方案异常的识别方法，其中，所述识别方法包括以下步骤：
    获取参保人的身份信息，并根据所述参保人的身份信息通过服务器从各个医疗机构或药店的管理系统中获取参保人在目标时间段内的购药明细以及所述购药明细所对应的处方明细；
    利用清洗模型对所述购药明细和所述处方明细进行清洗处理以得到对应的标准化字段，将所述购药明细的标准化字段与所述处方明细的标准化字段进行匹配，以判断所述购药明细中是否存在不属于所述处方明细的药物；
    若所述购药明细中存在不属于所述处方明细的药物，判断所述不属于所述处方明细的药物是否为处方药；
    若所述不属于所述处方明细的药物是处方药，则标记所述参保人存在用药方案异常的行为。
  2. 如权利要求1所述的用药方案异常的识别方法，其中，所述获取参保人的身份信息，并根据所述参保人的身份信息通过服务器从各个医疗机构或药店的管理系统中获取参保人在目标时间段内的购药明细以及所述购药明细所对应的处方明细的步骤，包括：
    根据参保人的身份信息通过服务器从各个医疗机构或药店的管理系统中获取参保人在目标时间段内的购药明细；
    根据第一预设规则获取目标购药明细所依据的处方明细的开具时间区间；
    根据所述开具时间区间获取所述参保人在各个医院的就诊记录；
    根据所述就诊记录获取所述参保人的处方明细。
  3. 如权利要求2所述的用药方案异常的识别方法，其中，所述根据第一预设规则获取目标购药明细所依据的处方明细的开具时间区间的步骤包括：
    获取在目标时间段内的最早购药行为发生之日和最晚购药行为发生之日；
    根据所述最早购药行为发生之日减去预设时间段得到最早开具之日，其中所述开具时间区间为所述最早开具之日至最晚购药行为发生之日组成的时间区间。
  4. 如权利要求2所述的用药方案异常的识别方法，其中，所述根据所述开具时间区间获取所述参保人在各个医院的就诊记录的步骤之后，还包括：
    根据所述就诊记录确定所述参保人的历史病种，并分析所述购药明细中的药物的对症病种；
    判断所述历史病种与所述对症病种是否相同；
    若所述历史病种与所述对症病种不相同，则获取所述历史病种对应的辅助用药范围，判断所述购药明细中是否存在属于所述辅助用药范围的药物；
    若所述购药明细中不存在属于所述辅助用药范围的药物，则标记所述参保人存在用药方案异常的行为。
  5. 如权利要求1所述用药方案异常的识别方法，其中，所述判断所述购药明细中是否存在不属于所述处方明细的药物的步骤，包括：
    按照第二预设规则依次选出多份所述处方明细中的一份处方明细作为目标处方明细；
    若所述购药明细中的药物全部属于所述目标处方明细，则判定所述购药明细中不存在不属于所述处方明细的药物；
    若所述购药明细中的药物部分或全部不属于所述目标处方明细，则判定所述购药明细中存在不属于所述处方明细的药物。
  6. 如权利要求5所述的用药方案异常的识别方法，其中，所述按照第二预设规则依次选出多份所述处方明细中的一份处方明细作为目标处方明细的步骤包括：
    按照第二预设规则从多份所述处方明细中选出第一目标处方明细，判断所述购药明细中是否存在不属于所述第一目标处方明细的药物；
    若所述购药明细中存在不属于所述第一目标处方明细的药物，则对所述购药明细中不属于所述第一目标处方明细的药物进行标记，并标记为待审药物；
    从剩余所述处方明细中选出剩余目标处方明细，并将所述待审药物与剩余目标处方明细记载的药物逐张进行对比。
  7. 如权利要求1所述的用药方案异常的识别方法，其中，所述若所述不属于所述处方明细的药物是处方药，则标记所述参保人存在用药方案异常的行为的步骤包括：
    若所述不属于所述处方明细的药物是处方药，则获取与所述参保人为共用医保账户关系的间接参保人的身份信息；
    根据所述间接参保人的身份信息和所述开具时间区间，获取各个医院的与所述间接参保人对应的间接处方明细；
    判断所述不属于所述处方明细的药物是否属于所述间接处方明细；
    若所述不属于参保人的所述处方明细的药物也不属于所述间接处方明细，则标记所述参保人存在用药方案异常的行为。
  8. 一种用药方案异常的识别装置，其中，包括：
    获取模块，用于获取参保人的身份信息，并根据所述参保人的身份信息通过服务器从各个医疗机构或药店的管理系统中获取参保人在目标时间段内的购药明细以及所述购药明细所对应的处方明细；
    判断模块，用于利用清洗模型对所述购药明细和所述处方明细进行清洗处理以得到对应的标准化字段，将所述购药明细的标准化字段与所述处方明细的标准化字段进行匹配，以判断所述购药明细中是否存在不属于所述处方明细的药物；
    标记模块，用于若所述购药明细中存在不属于所述处方明细的药物，则将所述参保人标记为存在用药方案异常的行为。
  9. 一种终端，其中，包括处理器、存储器、以及存储在所述存储器上并可被所述处理器执行的计算机可读指令，其中所述计算机可读指令被所述处理器执行时，实现如下步骤：
    获取参保人的身份信息，并根据所述参保人的身份信息通过服务器从各个医疗机构或药店的管理系统中获取参保人在目标时间段内的购药明细以及所述购药明细所对应的处方明细；
    利用清洗模型对所述购药明细和所述处方明细进行清洗处理以得到对应的标准化字段，将所述购药明细的标准化字段与所述处方明细的标准化字段进行匹配，以判断所述购药明细中是否存在不属于所述处方明细的药物；
    若所述购药明细中存在不属于所述处方明细的药物，判断所述不属于所述处方明细的药物是否为处方药；
    若所述不属于所述处方明细的药物是处方药，则标记所述参保人存在用药方案异常的行为。
  10. 如权利要求9所述的终端，其中，所述计算机可读指令被所述处理器执行时，还实现如下步骤：
    根据参保人的身份信息通过服务器从各个医疗机构或药店的管理系统中获取参保人在目标时间段内的购药明细；
    根据第一预设规则获取目标购药明细所依据的处方明细的开具时间区间；
    根据所述开具时间区间获取所述参保人在各个医院的就诊记录；
    根据所述就诊记录获取所述参保人的处方明细。
  11. 如权利要求10所述的终端，其中，所述计算机可读指令被所述处理器执行时，还实现如下步骤：
    获取在目标时间段内的最早购药行为发生之日和最晚购药行为发生之日；
    根据所述最早购药行为发生之日减去预设时间段得到最早开具之日，其中所述开具时间区间为所述最早开具之日至最晚购药行为发生之日组成的时间区间。
  12. 如权利要求10所述的终端，其中，所述计算机可读指令被所述处理器执行时，还实现如下步骤：
    根据所述就诊记录确定所述参保人的历史病种，并分析所述购药明细中的药物的对症病种；
    判断所述历史病种与所述对症病种是否相同；
    若所述历史病种与所述对症病种不相同，则获取所述历史病种对应的辅助用药范围，判断所述购药明细中是否存在属于所述辅助用药范围的药物；
    若所述购药明细中不存在属于所述辅助用药范围的药物，则标记所述参保人存在用药方案异常的行为。
  13. 如权利要求9所述的终端，其中，所述计算机可读指令被所述处理器执行时，还实现如下步骤：
    按照第二预设规则依次选出多份所述处方明细中的一份处方明细作为目标处方明细；
    若所述购药明细中的药物全部属于所述目标处方明细，则判定所述购药明细中不存在不属于所述处方明细的药物；
    若所述购药明细中的药物部分或全部不属于所述目标处方明细，则判定所述购药明细中存在不属于所述处方明细的药物。
  14. 如权利要求9所述的终端，其中，所述计算机可读指令被所述处理器执行时，还实现如下步骤：
    若所述不属于所述处方明细的药物是处方药，则获取与所述参保人为共用医保账户关系的间接参保人的身份信息；
    根据所述间接参保人的身份信息和所述开具时间区间，获取各个医院的与所述间接参保人对应的间接处方明细；
    判断所述不属于所述处方明细的药物是否属于所述间接处方明细；
    若所述不属于参保人的所述处方明细的药物也不属于所述间接处方明细，则标记所述参保人存在用药方案异常的行为。
  15. 一种可读存储介质，其中，所述可读存储介质上存储有计算机可读指令，其中所述计算机可读指令被处理器执行时，实现如下步骤：
    获取参保人的身份信息，并根据所述参保人的身份信息通过服务器从各个医疗机构或药店的管理系统中获取参保人在目标时间段内的购药明细以及所述购药明细所对应的处方明细；
    利用清洗模型对所述购药明细和所述处方明细进行清洗处理以得到对应的标准化字段，将所述购药明细的标准化字段与所述处方明细的标准化字段进行匹配，以判断所述购药明细中是否存在不属于所述处方明细的药物；
    若所述购药明细中存在不属于所述处方明细的药物，判断所述不属于所述处方明细的药物是否为处方药；
    若所述不属于所述处方明细的药物是处方药，则标记所述参保人存在用药方案异常的行为。
  16. 如权利要求15所述的计算机可读存储介质，其中，所述计算机可读指令被所述处理器执行时，还实现如下步骤：
    根据参保人的身份信息通过服务器从各个医疗机构或药店的管理系统中获取参保人在目标时间段内的购药明细；
    根据第一预设规则获取目标购药明细所依据的处方明细的开具时间区间；
    根据所述开具时间区间获取所述参保人在各个医院的就诊记录；
    根据所述就诊记录获取所述参保人的处方明细。
  17. 如权利要求16所述的计算机可读存储介质，其中，所述计算机可读指令被所述处理器执行时，还实现如下步骤：
    获取在目标时间段内的最早购药行为发生之日和最晚购药行为发生之日；
    根据所述最早购药行为发生之日减去预设时间段得到最早开具之日，其中所述开具时间区间为所述最早开具之日至最晚购药行为发生之日组成的时间区间。
  18. 如权利要求16所述的计算机可读存储介质，其中，所述计算机可读指令被所述处理器执行时，还实现如下步骤：
    根据所述就诊记录确定所述参保人的历史病种，并分析所述购药明细中的药物的对症病种；
    判断所述历史病种与所述对症病种是否相同；
    若所述历史病种与所述对症病种不相同，则获取所述历史病种对应的辅助用药范围，判断所述购药明细中是否存在属于所述辅助用药范围的药物；
    若所述购药明细中不存在属于所述辅助用药范围的药物，则标记所述参保人存在用药方案异常的行为。
  19. 如权利要求15所述的计算机可读存储介质，其中，所述计算机可读指令被所述处理器执行时，还实现如下步骤：
    按照第二预设规则从多份所述处方明细中选出第一目标处方明细，判断所述购药明细中是否存在不属于所述第一目标处方明细的药物；
    若所述购药明细中存在不属于所述第一目标处方明细的药物，则对所述购药明细中不属于所述第一目标处方明细的药物进行标记，并标记为待审药物；
    从剩余所述处方明细中选出剩余目标处方明细，并将所述待审药物与剩余目标处方明细记载的药物逐张进行对比。
  20. 如权利要求15所述的计算机可读存储介质，其中，所述计算机可读指令被所述处理器执行时，还实现如下步骤：
    若所述不属于所述处方明细的药物是处方药，则获取与所述参保人为共用医保账户关系的间接参保人的身份信息；
    根据所述间接参保人的身份信息和所述开具时间区间，获取各个医院的与所述间接参保人对应的间接处方明细；
    判断所述不属于所述处方明细的药物是否属于所述间接处方明细；
    若所述不属于参保人的所述处方明细的药物也不属于所述间接处方明细，则标记所述参保人存在用药方案异常的行为。
PCT/CN2019/097445 2018-12-13 2019-07-24 用药方案异常的识别方法、装置、终端及可读存储介质 WO2020119131A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201811531071.9A CN109636650A (zh) 2018-12-13 2018-12-13 用药方案异常的识别方法、装置、终端及可读存储介质
CN201811531071.9 2018-12-13

Publications (1)

Publication Number Publication Date
WO2020119131A1 true WO2020119131A1 (zh) 2020-06-18

Family

ID=66073922

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2019/097445 WO2020119131A1 (zh) 2018-12-13 2019-07-24 用药方案异常的识别方法、装置、终端及可读存储介质

Country Status (2)

Country Link
CN (1) CN109636650A (zh)
WO (1) WO2020119131A1 (zh)

Families Citing this family (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109636650A (zh) * 2018-12-13 2019-04-16 平安医疗健康管理股份有限公司 用药方案异常的识别方法、装置、终端及可读存储介质
CN110675958B (zh) * 2019-08-31 2023-03-31 天津开心生活科技有限公司 一种肿瘤治疗方案判线方法及装置
CN111028089B (zh) * 2019-11-25 2023-05-02 泰康保险集团股份有限公司 异常操作识别方法及装置、计算机存储介质、电子设备
CN112835893B (zh) * 2021-01-18 2023-03-21 浙江大学山东工业技术研究院 一种基于聚类的医保欺诈行为的检测方法及系统
CN113990436B (zh) * 2021-12-27 2022-03-01 西藏自治区人民政府驻成都办事处医院 基于矩阵校验的快速判断用药合理性的方法及系统
CN115831312A (zh) * 2022-11-24 2023-03-21 上海市精神卫生中心（上海市心理咨询培训中心） 一种用药异常识别方法及系统

Citations (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2011022978A (ja) * 2009-07-16 2011-02-03 Medi-Pharm:Kk 健康保険組合事業者の医療費削減事業
CN103514576A (zh) * 2013-09-06 2014-01-15 深圳民太安信息技术有限公司 一种社保就诊违规套现的筛查方法
US20160246942A1 (en) * 2015-01-29 2016-08-25 American Institute of Toxicology, Inc. D/b/a AIT Laboratories Narcotics monitoring system and method
CN106408481A (zh) * 2015-07-28 2017-02-15 平安科技（深圳）有限公司 异常刷卡人员信息自动提取系统及方法
CN107133437A (zh) * 2017-03-03 2017-09-05 平安医疗健康管理股份有限公司 监控药品使用的方法及装置
CN107133438A (zh) * 2017-03-03 2017-09-05 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
CN107451400A (zh) * 2017-07-11 2017-12-08 武汉金豆医疗数据科技有限公司 一种医疗行为监控方法和系统
CN108492196A (zh) * 2018-03-08 2018-09-04 平安医疗健康管理股份有限公司 通过数据分析推断医疗保险违规行为的风控方法
CN109636650A (zh) * 2018-12-13 2019-04-16 平安医疗健康管理股份有限公司 用药方案异常的识别方法、装置、终端及可读存储介质

Family Cites Families (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104182824B (zh) * 2014-08-08 2018-02-06 平安养老保险股份有限公司 一种识别医保报销违规行为的规则校验系统及校验方法
CN106126935A (zh) * 2016-06-24 2016-11-16 北京千安哲信息技术有限公司 一种安全用药监测系统及其监测方法
CN109545317A (zh) * 2018-10-30 2019-03-29 平安科技（深圳）有限公司 基于住院预测模型判定住院行为的方法及相关产品

Patent Citations (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2011022978A (ja) * 2009-07-16 2011-02-03 Medi-Pharm:Kk 健康保険組合事業者の医療費削減事業
CN103514576A (zh) * 2013-09-06 2014-01-15 深圳民太安信息技术有限公司 一种社保就诊违规套现的筛查方法
US20160246942A1 (en) * 2015-01-29 2016-08-25 American Institute of Toxicology, Inc. D/b/a AIT Laboratories Narcotics monitoring system and method
CN106408481A (zh) * 2015-07-28 2017-02-15 平安科技（深圳）有限公司 异常刷卡人员信息自动提取系统及方法
CN107133437A (zh) * 2017-03-03 2017-09-05 平安医疗健康管理股份有限公司 监控药品使用的方法及装置
CN107133438A (zh) * 2017-03-03 2017-09-05 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
CN107451400A (zh) * 2017-07-11 2017-12-08 武汉金豆医疗数据科技有限公司 一种医疗行为监控方法和系统
CN108492196A (zh) * 2018-03-08 2018-09-04 平安医疗健康管理股份有限公司 通过数据分析推断医疗保险违规行为的风控方法
CN109636650A (zh) * 2018-12-13 2019-04-16 平安医疗健康管理股份有限公司 用药方案异常的识别方法、装置、终端及可读存储介质

Also Published As

Publication number Publication date
CN109636650A (zh) 2019-04-16

Similar Documents

Publication Publication Date Title
WO2020119131A1 (zh) 用药方案异常的识别方法、装置、终端及可读存储介质
WO2020213843A1 (ko) 사용자 맞춤형 의료정보 제공 시스템 및 이의 구동방법
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
WO2020119176A1 (zh) 报销数据的排查方法、识别服务端及存储介质
WO2020078053A1 (zh) 医疗数据异常检测方法、装置、设备及存储介质
CN111145844B (zh) 综合医疗监管平台
WO2021141388A1 (ko) 건강사고의 발생확률을 예측하는 건강등급 평가 시스템 및 방법
WO2020119403A1 (zh) 住院数据异常检测方法、装置、设备及可读存储介质
US7725479B2 (en) Unique person registry
JP5745933B2 (ja) 病歴診断システムおよび方法
CA2564307C (en) Data record matching algorithms for longitudinal patient level databases
WO2020119385A1 (zh) 基于数据分析的处方生成监测方法、装置、设备和介质
WO2020119401A1 (zh) 医师违规执业的监控方法、监控服务端及存储介质
US20080120296A1 (en) Systems and methods for free text searching of electronic medical record data
WO2020108111A1 (zh) 医保欺诈行为的识别方法、装置、设备及可读存储介质
WO2022145782A2 (ko) 빅데이터 및 클라우드 시스템 기반 인공지능 응급의료 의사결정 및 응급환자 이송 시스템과 그 방법
WO2020078058A1 (zh) 医疗数据异常识别方法、装置、终端及存储介质
WO2020226456A1 (ko) 처방 정보를 통한 의료 정보 제공 방법 및 장치
Khaleghi et al. New ways to manage pandemics: using technologies in the era of COVID-19: a narrative review
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
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
WO2016165177A1 (zh) 基于面部识别的网络医院自助缴费方法和系统
WO2016165201A1 (zh) 基于网络医院的自助缴费方法和系统
WO2019100595A1 (zh) 保险理赔案件互斥信息的确定方法及装置
WO2022035018A1 (ko) 클리닉 매니저 서비스 플랫폼 및 이를 제공하는 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: 19895911

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

32PN Ep: public notification in the ep bulletin as address of the adressee cannot be established

Free format text: NOTING OF LOSS OF RIGHTS PURSUANT TO RULE 112(1) EPC (EPO FORM 1205A DATED 15/10/2021)

122 Ep: pct application non-entry in european phase

Ref document number: 19895911

Country of ref document: EP

Kind code of ref document: A1