WO2020114221A1 - 基于生物特征识别的定点医药机构的监管方法及相关设备 - Google Patents
基于生物特征识别的定点医药机构的监管方法及相关设备 Download PDFInfo
- Publication number
- WO2020114221A1 WO2020114221A1 PCT/CN2019/118818 CN2019118818W WO2020114221A1 WO 2020114221 A1 WO2020114221 A1 WO 2020114221A1 CN 2019118818 W CN2019118818 W CN 2019118818W WO 2020114221 A1 WO2020114221 A1 WO 2020114221A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- medical insurance
- card
- biometric information
- face
- insurance card
- Prior art date
Links
Classifications
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H40/00—ICT specially adapted for the management or administration of healthcare resources or facilities; ICT specially adapted for the management or operation of medical equipment or devices
-
- 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
- G06Q20/00—Payment architectures, schemes or protocols
- G06Q20/38—Payment protocols; Details thereof
- G06Q20/40—Authorisation, e.g. identification of payer or payee, verification of customer or shop credentials; Review and approval of payers, e.g. check credit lines or negative lists
- G06Q20/401—Transaction verification
- G06Q20/4014—Identity check for transactions
- G06Q20/40145—Biometric identity checks
-
- 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
- G06V40/16—Human faces, e.g. facial parts, sketches or expressions
- G06V40/168—Feature extraction; Face representation
Definitions
- This application relates to the technical field of medical supervision, in particular to a supervision method and related equipment for designated medical institutions based on biometric identification.
- Social medical insurance is a social insurance system established by the state and society in accordance with certain laws and regulations to provide workers within the scope of protection with basic medical needs protection in case of illness.
- relevant government medical insurance management departments such as medical insurance centers
- Insured personnel of the medical insurance card can purchase medicines that are allowed to be paid with the medical insurance card at designated pharmacies, which makes it easier for the insured personnel to choose the designated pharmacies closer to them to purchase medicine and seek medical treatment.
- designated pharmacies use regulatory loopholes to conduct illegal card swiping (medical insurance cards), cash out, substitute payment, randomly raise drug prices, and sub-goods in order to benefit themselves. This not only violates the relevant regulations of the medical insurance management, makes the medical insurance fund unreasonable, and cannot provide essential and reasonable basic medical services for the insured, but also causes vicious competition in the industry and seriously affects the market order.
- the main purpose of this application is to provide a monitoring method and related equipment for designated medical institutions based on biometric identification, which aims to solve the problem that the designated medical institutions cannot be effectively supervised in the prior art, so that the designated medical institutions cannot use the regulatory loopholes Carry out the technical problems of illegal card swiping and arbitrary increase of drug prices.
- the present application provides a method for supervising a designated medical institution based on biometric identification.
- the method includes the following steps:
- the designated medical institution supervision platform responds to the card swiping request triggered by the card swiping device in the designated medical institution, and obtains the first biometric information of the user using the medical insurance card and the identification number of the medical insurance card;
- the identification number obtain the second biometric information of the holder of the medical insurance card corresponding to the identification number from the social security platform;
- the present application also proposes a monitoring device for designated medical institutions based on biometric identification, the device includes:
- the first obtaining module is used to obtain the first biometric information of the user using the medical insurance card and the identification number of the medical insurance card in response to the card swiping request triggered by the card swiping device in the designated medical institution.
- the second obtaining module is used to obtain the second biometric information of the holder of the medical insurance card corresponding to the identification number from the social security platform according to the identification number;
- the matching module is used to match the first biometric information with the second biometric information
- the sending module is used to issue an instruction prohibiting payment by the medical insurance card to the card swiping device when the first biometric information does not match the second biometric information, so that the card swiping device cannot deduct a fee from the medical insurance card.
- the present application also proposes a monitoring device for a designated medical institution based on biometric identification.
- the device includes: a memory, a processor, and computer-readable storage instructions stored on the memory and executable on the processor When the computer-readable instructions are executed by the processor, the steps of the monitoring method of the designated medical institution based on biometrics identification as described above are implemented.
- the present application also proposes a computer-readable storage medium having computer-readable instructions stored on the computer-readable storage medium, the computer-readable instructions executed by the processor are implemented as described above based on biometrics Identify the steps of the regulatory methods of designated medical institutions.
- the first biometric of the user who currently uses the medical insurance card to purchase the medicine and the medical insurance are acquired The identification number of the card, and obtain the second biometric information of the true holder of the medical insurance card corresponding to the current identification number from the social security platform according to the obtained identification number, and then the first biometric information and the second biometric information
- the feature information is matched, and the medical insurance card can only be used to pay for the purchased medicines when the two match, otherwise an instruction prohibiting payment using the medical insurance card is issued to the card swiping device that initiated the card swiping request, making the card swiping device unable
- the medical insurance card is successfully deducted, which can effectively prevent medical designated medical institutions from using regulatory loopholes to conduct illegal card swiping.
- FIG. 1 is a schematic structural diagram of a monitoring device of a designated medical institution based on biometric identification in a hardware operating environment involved in an embodiment of the present application;
- FIG. 2 is a schematic flowchart of a first embodiment of a method for supervising a designated medical institution based on biometric identification in this application;
- FIG. 3 is a schematic diagram of splitting a convolution kernel of size in a training model into two convolution kernels of size in a first embodiment of a supervision method of a designated medical institution based on biometric recognition of this application;
- FIG. 4 is a schematic diagram of splitting the convolution kernel of size in the training model into four convolution kernels of size in the first embodiment of the monitoring method of a designated medical institution based on biometric identification in this application;
- FIG. 5 is a schematic flowchart of a second embodiment of a method for supervising a designated medical institution based on biometric identification in this application;
- FIG. 6 is a structural block diagram of a first embodiment of a monitoring device for a designated medical institution based on biometric identification in this application.
- FIG. 1 is a schematic structural diagram of a supervised device of a designated medical institution based on biometric identification in a hardware operating environment according to an embodiment of the present application.
- the monitoring device of the designated medical institution based on biometric identification may include: a processor 1001, such as a central processor (Central Processing) Unit, CPU), communication bus 1002, user interface 1003, network interface 1004, memory 1005.
- the communication bus 1002 is used to implement connection communication between these components.
- the user interface 1003 may include a display (Display), an input unit such as a keyboard (Keyboard), and the optional user interface 1003 may also include a standard wired interface and a wireless interface.
- the network interface 1004 may optionally include a standard wired interface and a wireless interface (such as a wireless fidelity (WIreless-FIdelity, WI-FI) interface).
- the memory 1005 may be a high-speed random access memory (Random Access Memory (RAM) memory can also be a stable non-volatile memory (Non-Volatile Memory, NVM), such as disk storage.
- RAM Random Access Memory
- NVM Non-Volatile Memory
- the memory 1005 may optionally be a storage device independent of the foregoing processor 1001.
- FIG. 1 does not constitute a limitation on the supervision equipment of a designated medical institution based on biometric identification, and may include more or less components than the illustration, or a combination of certain components , Or different component arrangements.
- the memory 1005 as a computer-readable storage medium may include an operating system, a data storage module, a network communication module, a user interface module, and a supervisory program for a designated medical institution based on biometric identification.
- the network interface 1004 is mainly used for data communication with various devices of the designated medical institutions, such as card swiping equipment and image acquisition equipment; the user interface 1003 is mainly used For data interaction with users; the processor 1001 and the memory 1005 in the monitoring device of the designated medical institution based on biometric identification of this application may be set in the monitoring device of the designated medical institution based on biometric identification, based on the fixed point of biometric identification
- the supervisory device of the medical institution calls the supervisory program of the designated medical institution based on biometric identification stored in the memory 1005 through the processor 1001, and executes the supervisory method of the designated medical institution based on biometric identification provided in the embodiments of the present application.
- FIG. 2 is a schematic flowchart of a first embodiment of a method for supervising a designated medical institution based on biometric identification.
- the monitoring method for designated medical institutions based on biometric identification includes the following steps:
- Step S10 The designated medical institution supervision platform obtains the first biometric information of the user using the medical insurance card and the identification number of the medical insurance card in response to the card swiping request triggered by the card swiping device in the designated medical institution.
- the designated medical institution supervision platform provided in this case needs to be communicated with designated medical institutions in advance, such as designated chain pharmacies, designated monomer pharmacies, and designated medical institutions (such as designated hospitals).
- designated medical institutions such as designated chain pharmacies, designated monomer pharmacies, and designated medical institutions (such as designated hospitals).
- card-swapping devices such as special card-swapping machines, smart mobile devices with corresponding payment applications, such as mobile phones, tablets, etc., and various graphics collection devices, for medical staff to place orders.
- the terminal equipment of the pharmacy is in communication, so that data and instructions can be interacted with the above equipment.
- the designated medical institution supervision platform also needs to communicate with the social security platform, that is, the existing social security management center in advance, so that you can easily get the information of the insurance participants.
- the above-mentioned first biometric information may specifically be facial features, fingerprint feature information, iris feature information, voiceprint feature information, etc., which are not listed here one by one, nor are there any restrictions on this.
- a person skilled in the art may set the first biometric information to be obtained as needed.
- the selected device for acquiring the first biometric information will also be different.
- the first biometric information is facial feature information or iris feature information
- image acquisition is required.
- Device to obtain when the first biometric information is fingerprint characteristic information, you need to use the fingerprint module to obtain; when the first biometric information is voiceprint characteristic information, you need to use a voice device to obtain.
- the aforementioned medical insurance card that is, the social medical insurance card
- the social medical insurance card is also called a medical insurance card.
- it can be a social security card integrating social security and medical insurance.
- the above identification number is specifically an ID (Identification) number that identifies the uniqueness of the medical insurance card, or the ID number of the insured person.
- the following uses the first biometric information as the first facial feature information of the user using the medical insurance card, and the second biometric information as the medical
- the second face feature information of the holder of the insurance card will be specifically described as an example.
- acquiring the first biometric information of the user using the medical insurance card includes: (1) receiving an image from a designated medical institution Real-time streaming media file containing the first face image of the user using the medical insurance card uploaded by the collection device; (2) Perform face recognition on each frame of streaming media data in the real-time streaming media file according to the pre-stored face detection model After detection, a first face image is obtained; (3) According to a pre-stored face feature extraction model, face feature extraction is performed on the first face image to obtain first face feature information.
- the above-mentioned face detection model can be obtained by training the face sample data based on the convolution neural network algorithm, and the face feature extraction model can be based on the face sample data. Face feature training.
- the convolutional neural network is a deep feed-forward artificial neural network in machine learning, which can more accurately identify the information in the image. Therefore, through the convolutional neural network training on the face sample data (such as downloading from the network in advance, or shooting the recorded face image), a face detection model that can accurately identify the image is obtained.
- the face detection model used is specifically Caffe (Convolutional Neural Network Framework, Convolutional Architecture for Fast Feature Embedding), and then trained based on convolutional neural network algorithm. Since Caffe is a clear, highly readable and fast deep learning framework, training a face detection model based on Caffe can greatly improve the running speed and greatly reduce the size of the face detection model obtained by training. Make the trained face detection model a real-time target detection model, that is, whether the image acquisition device uploads a static image or a real-time streaming media file, based on the face detection model, the first user of the medical insurance card can be extracted from it Face image.
- Caffe Convolutional Neural Network Framework, Convolutional Architecture for Fast Feature Embedding
- the face detection model can also be obtained by convolutional neural network training of the real-time rapid target detection (YOLO) neural network model using a face recognition database with rich existing face data.
- YOLO real-time rapid target detection
- the upgraded version of YOLO can be used to obtain the convolutional neural network training.
- the extraction accuracy of the trained face feature extraction model can be greatly improved.
- This embodiment does the existing training method.
- the following improvement is specifically to split the convolution kernel in the training model before training.
- a training model is constructed based on the face features in the face sample data.
- sample data mentioned here may be face images downloaded from major data platforms in advance using the network, or face images captured and recorded in advance, which will not be listed one by one this time. Make no restrictions.
- the convolution kernel of size in the training model is split into at least two convolution kernels of size.
- the letter A in Figure 3 represents the convolution kernel of size in the training model, and the letters A1 and A2 are the four convolution kernels of size after splitting; the letter B in Figure 4 represents the In the training model, the size of the convolution kernel, the letters B1, B2, B3, and B4 are the four convolution kernels of size.
- the training model is trained to obtain a face feature extraction model.
- the face feature extraction model in this embodiment is also a convolutional neural network model, that is, the face feature extraction model is mainly composed of a convolution layer, a pooling layer, and a fully connected layer.
- the combination of the accumulation layer and the pooling layer can appear multiple times.
- the fully connected layer is located behind the pooling layer and serves as the output layer of the entire model.
- the facial features mentioned in this embodiment specifically use the output of the nodes in the output layer of the facial feature extraction model as the facial features.
- the facial features may be composed of various feature points of the face, such as eyes , The tip of the nose, the corner of the mouth, the eyebrows, and the contour points of other parts of the face.
- the face feature extraction model used in this embodiment can be normalized to the face sample data before training according to the face feature training in the face sample data, so that the normalized Face sample data can control the scale of each feature in the same range, thereby greatly reducing the number of nodes in the convolution kernel of each convolutional layer and the fully connected layer as the output layer during training, simplifying training
- the various calculations in the process can also improve the accuracy of the face feature extraction model constructed.
- Step S20 Obtain the second biometric information of the holder of the medical insurance card corresponding to the identification number from the social security platform according to the identification number.
- the holder of the medical insurance card mentioned above refers to the actual insured person corresponding to the medical insurance card. Therefore, according to the identification number, the second biometric information of the holder of the medical insurance card corresponding to the identification number is obtained from the social security platform, which is the face characteristic and fingerprint characteristic information entered by the insured person when applying for the medical insurance card. , Iris feature information, voiceprint feature information, etc.
- Step S30 Match the first biometric information with the second biometric information.
- the first facial feature information and the second facial feature information are performed
- the specific operation of matching is as follows: First, match the first face feature information with the second face feature information one by one to determine the cosine similarity between the first face feature information and the second face feature information; then, Compare the cosine similarity with the preset similarity threshold.
- Step S40 If the first biometric information and the second biometric information do not match, an instruction prohibiting payment by the medical insurance card is issued to the card swiping device, so that the card swiping device cannot deduct a fee from the medical insurance card.
- first biometric information as the first facial feature information
- second biometric information as the second facial feature information
- an instruction prohibiting payment using the medical insurance card is issued to the card swiping device.
- each type of information needs to be entered, and the corresponding quantity, purchase invoice, price set by the relevant department, etc., and the content of the entry can be as much as possible.
- the details are in order to facilitate later supervision and inquiry.
- the medical staff of the designated medical institution uses the terminal device to first check whether there are medicines selected by the user in the medicine library. Or, if the quantity has fallen below a certain threshold, a reminder will be promptly given to recommend other suitable alternative medicines according to the user's situation. If so, during the payment process, the above steps S10 to S40 are executed according to the flow.
- the designated medical institution supervision platform can also obtain the information of the medical personnel of each designated medical institution in advance, and check whether the medical personnel are qualified to practice medicine, and can prescribe, so as to avoid designated medical institutions from randomly issuing prescriptions when there are no doctors who can prescribe. .
- each designated medical institution has a doctor who has prescribed a prescription
- other personnel can prescribe at will, and the supervision of the doctor's appointment time can also be set, such as the use of various anti-punch equipment, making the doctor necessary Personal punch card, so as to avoid others from randomly writing prescriptions when the doctor is not on duty.
- Biometrics and the identification number of the medical insurance card and obtain the second biometric information of the true holder of the medical insurance card corresponding to the current identification number from the social security platform according to the obtained identification number, and then the first biological
- the feature information is matched with the second biometric information, and the medical insurance card can only be used to pay for the purchased medicines when the two match, otherwise an instruction to prohibit payment using the medical insurance card is issued to the card swiping device that initiated the card swiping request, This prevents the card swiping device from successfully deducting fees from the medical insurance card, and thus can effectively prevent medical designated medical institutions from using regulatory loopholes to conduct illegal card swiping.
- FIG. 5 is a schematic flowchart of a second embodiment of a monitoring method for a designated medical institution based on biometric identification. Based on the first embodiment described above, the method for supervising a designated medical institution based on biometric identification in this embodiment after step S30, further includes:
- Step S50 It is determined that the medicine can be purchased using the medical insurance card, and an instruction allowing payment by the medical insurance card is issued to the card swiping device, so that the card swiping device deducts the fee from the medical insurance card.
- step S50 In order to facilitate understanding of the implementation process of the above step S50, the following is a specific description:
- step S30 if it is determined that the first biometric information matches the second biometric information by executing step S30, first, the information of the medicine purchased by the user using the medical insurance card is acquired.
- the information of the drugs mentioned here can specifically include the name, number, manufacturer, type of drug, and project specifications (such as how many bags per box, how many grams per bag, etc.), etc.
- project specifications such as how many bags per box, how many grams per bag, etc.
- the pre-stored basic medical insurance drug catalog and drug information determine whether the drug can be purchased using the medical insurance card, if the drug can be purchased using the medical insurance card, then issue the instruction to allow the payment using the medical insurance card to the card swiping device, to Make the credit card deduction from the medical insurance card.
- the above basic medical insurance drug catalog may be stored in the form of Table 1.
- Table 1 Storage table of the basic medical insurance drug catalog Numbering Drug Name Invoice name Manufacturer Drug category Project specifications unit Whether medical insurance Deductible ratio 20102049073321910101 Compound Ganmaoling Granules Medicine fee Guangxi Baorita Pharmaceutical Co., Ltd.
- the monitoring method for designated medical institutions based on biometric identification matches the first biometric information with the second biometric information, and 2.
- biometric information matches obtain the information of the medicine purchased by the user using the medical insurance card, and determine whether the medicine can be purchased using the medical insurance card based on the pre-stored basic medical insurance medicine catalog and the information of the medicine. Only when the medical insurance card is purchased, the instruction to allow payment using the medical insurance card is issued to the card swiping device, which can further avoid designated medical institutions to use non-medical insurance drugs instead of medical insurance drugs and defraud medical insurance money, which not only causes losses to the insured, It also led to the unreasonable use of medical insurance funds.
- the computer-readable instructions may be stored in In a computer-readable storage medium, the aforementioned computer-readable storage medium may be a read-only memory, a magnetic disk, or an optical disk.
- an embodiment of the present invention also proposes a computer-readable storage medium having computer-readable instructions stored on the computer-readable storage medium. When the computer-readable instructions are executed by a processor, the biological-based storage medium described above is implemented. Steps of the method of supervising the designated medical institutions for feature recognition.
- the computer-readable storage medium may be a non-volatile computer-readable storage medium.
- FIG. 6 is a structural block diagram of a first embodiment of a monitoring device for a designated medical institution based on biometric identification of the present application.
- the monitoring device for a designated medical institution based on biometric identification provided by an embodiment of the present application includes: a first acquisition module 6001, a second acquisition module 6002, a matching module 6003, and a sending module 6004.
- the first obtaining module 6001 is configured to obtain the first biometric information of the user using the medical insurance card and the identification number of the medical insurance card in response to the card swiping request triggered by the card swiping device in the designated medical institution.
- the second obtaining module 6002 is used to obtain the second biometric information of the holder of the medical insurance card corresponding to the identification number from the social security platform according to the identification number; the matching module 6003 is used to combine the first biometric information with the second Biometric information matching; the sending module 6004 is used to issue an instruction prohibiting payment with a medical insurance card to the credit card device when the first biometric information does not match the second biometric information, so that the credit card device cannot obtain medical insurance Fees are deducted from the card.
- the first biometric information acquired by the first acquisition module 6001 may specifically be the first facial feature information of a user using a medical insurance card
- the second biometric information acquired by the second acquisition module 6002 The second biometric information may specifically be the second facial feature information of the holder of the medical insurance card. Therefore, the above-mentioned first obtaining module 6001 can be specifically refined into a real-time streaming media file receiving sub-module, a face detecting sub-module and a face feature information extracting sub-module.
- the operation of the first acquiring module 6001 to acquire the first biometric information of the user who uses the medical insurance card is as follows: First, the streaming media file receiving submodule , Receive the real-time streaming media file containing the first face image of the user using the medical insurance card uploaded by the image acquisition device in the designated medical institution; then, the face detection sub-module, according to the pre-stored face detection model, the real-time streaming Each frame of streaming media data in the media file is subjected to face detection to obtain the first face image; finally, the face feature extraction sub-module performs face detection on the first face image according to the pre-stored face feature extraction model Feature extraction to get the first face feature information.
- the matching module 6003 matches the first biometric information with the second biometric information, specifically: The two face features are matched.
- the above-mentioned face detection model can be obtained by training the face sample data based on the convolution neural network algorithm, and the face feature extraction model can be based on the face sample data. Face feature training.
- the supervision device of the designated medical institution based on biometric recognition may further include a face detection model construction module and a face feature extraction model construction module.
- the convolutional neural network Asing the convolutional neural network, those skilled in the art can know that it is a deep feed-forward artificial neural network in machine learning, which can more accurately identify the information in the image. Therefore, through the convolutional neural network training on the face sample data (such as downloading from the network in advance, or shooting the recorded face image), a face detection model that can accurately identify the image is obtained. Regarding its specific training process, a person skilled in the art can achieve this by searching for relevant materials, which will not be repeated here.
- this embodiment makes an improvement to the existing training method. Specifically, before training, the convolution kernel in the training model is first split. For ease of understanding, the following The operations performed by the face feature extraction model construction module are described in detail: first, the training model is constructed based on the face features in the face sample data; then, the convolution kernel of size in the training model is split into at least two sizes of Convolution kernel; Finally, based on the convolutional neural network algorithm, the training model is trained to obtain a face feature extraction model.
- the device may further include a normalization processing module, so that before the face feature extraction model construction module constructs a training model based on the face features in the face sample data, the normalization processing module first normalizes the face sample data The normalization process enables the normalized face sample data to control the scale of each feature in the same range, thereby improving the accuracy of the constructed face feature extraction model.
- the first biological characteristics of the user who is currently using the medical insurance card to purchase medicines and the identification number of the medical insurance card are obtained, and according to The obtained identification number obtains the second biometric information of the real holder of the medical insurance card corresponding to the current identification number from the social security platform, and then matches the first biometric information with the second biometric information.
- the medical insurance card Only when matching, can the medical insurance card be used to pay for the cost of the purchased medicines, otherwise the card swiping device that initiated the card swiping request will be issued an instruction prohibiting payment by the medical insurance card, so that the card swiping device cannot successfully deduct the fee from the medical insurance card Therefore, it can effectively prevent medical designated medical institutions from using regulatory loopholes to conduct illegal card swiping.
- the monitoring device of the designated medical institution based on biometric identification further includes a medicine information acquisition module, a medical insurance medicine judgment module, and a second sending module.
- the medicine information acquisition module is used to obtain information about medicines purchased by users who use the medical insurance card when the first biometric information matches the second biometric information
- the medical insurance medicine judgment module is used to base on the pre-stored basic medical insurance Drug catalog and drug information to determine whether a drug can be purchased using a medical insurance card
- the second sending module is used to issue instructions to the credit card device to allow payment using a medical insurance card when the drug can be purchased using a medical insurance card, so that Credit card equipment is deducted from the medical insurance card.
- the instruction to pay with a medical insurance card can further prevent designated medical institutions from using non-medical insurance drugs instead of medical insurance drugs and defrauding medical insurance money, which not only causes losses to the insured, but also leads to the unreasonable use of medical insurance funds.
- the monitoring device may also include: a medical insurance drug price reasonableness judgment module, so that the medical insurance drug price can be determined by the medical insurance card before the medical insurance drug judgment module determines that the drug can be purchased with the medical insurance card, and the credit card device is allowed to pay the instruction using the medical insurance card.
- the rationality judgment module obtains the medicine purchase order corresponding to the medicine that can be purchased with the medical insurance card, and extracts the price of the medicine from the medicine purchase order, and then judges whether the medicine price meets the medical insurance medicine according to the pre-stored medical insurance medicine cost standard and the medicine price
- the provisions of the cost standard which can control the second sending module that issues the instruction to allow the use of the medical insurance card to the card swiping device. Only when the medicine can be purchased with the medical insurance card and the price of the medicine meets the requirements of the medical insurance drug cost standard, the card is swiped The device issued instructions to allow the use of medical insurance cards. In this way, the effective supervision of designated medical institutions can be achieved, so that the medical insurance fund can be used reasonably, and the market order is guaranteed, which in turn better protects the interests of insured persons.
- the method of the embodiment can be implemented by means of software plus the necessary general hardware platform, and of course it can also be implemented by hardware, but in many cases the former is the better implementation.
- the technical solution can be embodied in the form of a software product in essence or part of the contribution to the existing technology.
- the computer software product is stored in a computer-readable storage medium (such as read-only memory (Read Only) Memory, ROM)/RAM, disk, optical Disk), including several instructions to enable a terminal device (which may be a mobile phone, a computer, a server, or a network device, etc.) to execute the method of each embodiment of the present application.
- a terminal device which may be a mobile phone, a computer, a server, or a network device, etc.
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Health & Medical Sciences (AREA)
- Accounting & Taxation (AREA)
- Physics & Mathematics (AREA)
- Finance (AREA)
- General Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- Theoretical Computer Science (AREA)
- General Health & Medical Sciences (AREA)
- Strategic Management (AREA)
- Oral & Maxillofacial Surgery (AREA)
- Computer Security & Cryptography (AREA)
- Technology Law (AREA)
- Human Computer Interaction (AREA)
- Multimedia (AREA)
- Development Economics (AREA)
- Economics (AREA)
- Marketing (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Biomedical Technology (AREA)
- Epidemiology (AREA)
- Medical Informatics (AREA)
- Primary Health Care (AREA)
- Public Health (AREA)
- Collating Specific Patterns (AREA)
- Medical Treatment And Welfare Office Work (AREA)
Abstract
一种基于生物特征识别的定点医药机构的监管方法及相关设备。该方法包括:定点医药机构监管平台响应于刷卡设备触发的使用医疗保险卡支付的刷卡请求,获取使用医疗保险卡的用户的第一生物特征信息和医疗保险卡的识别号(S10);根据识别号,从社会保障平台获取识别号对应的医疗保险卡的持有者的第二生物特征信息(S20);将第一生物特征信息与第二生物特征信息进行匹配(S30);若第一生物特征信息与第二生物特征信息不匹配,则向刷卡设备下发禁止使用医疗保险卡支付的指令(S40)。
Description
本申请要求于2018年12月4日提交中国专利局、申请号为201811473507.3发明名称为“基于生物特征识别的定点医药机构的监管方法及相关设备”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
本申请涉及医药监管技术领域,尤其涉及一种基于生物特征识别的定点医药机构的监管方法及相关设备。
背景技术
社会医疗保险是国家和社会根据一定的法律法规,为向保障范围内的劳动者提供患病时基本的医疗需求保障而建立的社会保险制度。为了方便参保人员就医、购药,政府医疗保险管理有关部门(如医保中心)根据国家规定,按照一定的标准从众多社会药店中严格遴选出一批符合条件的社会药店作为定点药店,即持医疗保险卡的参保人员在定点药店可以购买允许使用医疗保险卡支付的药品,从而方便参保人员选择距离自己较近的定点药店购药和就医。并且,为了进一步方便参保人员就医、购药,在遴选定点药店时需要遵循的“两定资格审查”也逐渐取消,放宽了对定点药店的审核,使得定点药店的数量越来越多,参保人员能够更加方便的就医和购药。
但是,由于目前没有较为完整的监管政策,因而随着定点药店的增加,问题也接踵而至。比如,定点药店为了自身利益,利用监管漏洞进行违规刷卡(医疗保险卡)、套现、代刷、随意抬高药价、以次充好的现象层出不穷。这不仅违反了医保管理相关规定,使得医疗保险基金不能合理使用,不能为参保人提供必要、合理的基本医疗服务,同时也造成了行业内的恶性竞争,严重影响市场秩序。
所以,亟需提供一种能够对定点药店等定点医药机构进行监管,以防止定点医药机构利用监管漏洞进行违规刷卡、随意抬高药价的方法。。
发明内容
本申请的主要目的在于提供一种基于生物特征识别的定点医药机构的监管方法及相关设备,旨在解决现有技术中无法对定点医药机构进行有效的监管,从而无法避免定点医药机构利用监管漏洞进行违规刷卡、随意抬高药价的技术问题。
为实现上述目的,本申请提供了一种基于生物特征识别的定点医药机构的监管方法,所述方法包括以下步骤:
定点医药机构监管平台响应于定点医药机构中刷卡设备触发的使用医疗保险卡支付的刷卡请求,获取使用医疗保险卡的用户的第一生物特征信息和医疗保险卡的识别号;
根据识别号,从社会保障平台获取识别号对应的医疗保险卡的持有者的第二生物特征信息;
将第一生物特征信息与第二生物特征信息进行匹配;
若第一生物特征信息与第二生物特征信息不匹配,则向刷卡设备下发禁止使用医疗保险卡支付的指令,以使刷卡设备无法从医疗保险卡中扣费。
此外,为实现上述目的,本申请还提出一种基于生物特征识别的定点医药机构的监管装置,所述装置包括:
第一获取模块,用于响应于定点医药机构中刷卡设备触发的使用医疗保险卡支付的刷卡请求,获取使用医疗保险卡的用户的第一生物特征信息和医疗保险卡的识别号;
第二获取模块,用于根据识别号,从社会保障平台获取识别号对应的医疗保险卡的持有者的第二生物特征信息;
匹配模块,用于将第一生物特征信息与第二生物特征信息进行匹配;
发送模块,用于在第一生物特征信息与第二生物特征信息不匹配时,向刷卡设备下发禁止使用医疗保险卡支付的指令,以使刷卡设备无法从医疗保险卡中扣费。
此外,为实现上述目的,本申请还提出一种基于生物特征识别的定点医药机构的监管设备,设备包括:存储器、处理器及存储在存储器上并可在处理器上运行的计算机可读存储指令,所述计算机可读指令被处理器执行时实现如上文的基于生物特征识别的定点医药机构的监管方法的步骤。
此外,为实现上述目的,本申请还提出一种计算机可读存储介质,计算机可读存储介质上存储有计算机可读指令,所述计算机可读指令被处理器执行时实现如上文的基于生物特征识别的定点医药机构的监管方法的步骤。
本实施例的基于生物特征识别的定点医药机构的监管方法及相关设备,在有用户使用医疗保险卡购买药品时,通过获取当前使用医疗保险卡购买药品的用户的第一生物特征和该医疗保险卡的识别号,并根据获取到的识别号从社会保障平台获取与当前识别号对应的医疗保险卡的真正的持有者的第二生物特征信息,然后将第一生物特征信息与第二生物特征信息进行匹配,在二者匹配时才能利用该医疗保险卡支付购买的药品的费用,否则便向发起刷卡请求的刷卡设备下发禁止使用该医疗保险卡支付的指令,使得刷卡设备无法从该医疗保险卡中成功扣费,从而可以有效的避免医疗定点医药机构利用监管漏洞进行违规刷卡。
附图说明
图1是本申请实施例方案涉及的硬件运行环境的基于生物特征识别的定点医药机构的监管设备的结构示意图;
图2为本申请基于生物特征识别的定点医药机构的监管方法第一实施例的流程示意图;
图3为本申请基于生物特征识别的定点医药机构的监管方法第一实施例中将训练模型中尺寸为的卷积核拆分为两个尺寸为的卷积核的示意图;
图4为本申请基于生物特征识别的定点医药机构的监管方法第一实施例中将训练模型中尺寸为的卷积核拆分为四个尺寸为的卷积核的示意图;
图5为本申请基于生物特征识别的定点医药机构的监管方法第二实施例的流程示意图;
图6为本申请基于生物特征识别的定点医药机构的监管装置第一实施例的结构框图。
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用以解释本申请,并不用于限定本申请。
参照图1,图1为本申请实施例方案涉及的硬件运行环境的基于生物特征识别的定点医药机构的监管设备结构示意图。
如图1所示,该基于生物特征识别的定点医药机构的监管设备可以包括:处理器1001,例如中央处理器(Central Processing
Unit,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),可选用户接口1003还可以包括标准的有线接口、无线接口。网络接口1004可选的可以包括标准的有线接口、无线接口(如无线保真(WIreless-FIdelity,WI-FI)接口)。存储器1005可以是高速的随机存取存储器(Random
Access Memory,RAM)存储器,也可以是稳定的非易失性存储器(Non-Volatile
Memory,NVM),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中示出的结构并不构成对基于生物特征识别的定点医药机构的监管设备的限定,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1所示,作为一种计算机可读存储介质的存储器1005中可以包括操作系统、数据存储模块、网络通信模块、用户接口模块以及基于生物特征识别的定点医药机构的监管程序。
在图1所示的基于生物特征识别的定点医药机构的监管设备中,网络接口1004主要用于与定点医药机构的各种设备,如刷卡设备、图像采集设备进行数据通信;用户接口1003主要用于与用户进行数据交互;本申请基于生物特征识别的定点医药机构的监管设备中的处理器1001、存储器1005可以设置在基于生物特征识别的定点医药机构的监管设备中,基于生物特征识别的定点医药机构的监管设备通过处理器1001调用存储器1005中存储的基于生物特征识别的定点医药机构的监管程序,并执行本申请实施例提供的基于生物特征识别的定点医药机构的监管方法。
本申请实施例提供了一种基于生物特征识别的定点医药机构的监管方法,参照图2,图2为本申请一种基于生物特征识别的定点医药机构的监管方法第一实施例的流程示意图。本实施例中,基于生物特征识别的定点医药机构的监管方法包括以下步骤:
步骤S10:定点医药机构监管平台响应于定点医药机构中刷卡设备触发的使用医疗保险卡支付的刷卡请求,获取使用医疗保险卡的用户的第一生物特征信息和医疗保险卡的识别号。
具体的说,为了保证该监管方法的顺利执行,本案中提供的定点医药机构监管平台需要预先与各定点医药机构,如定点连锁药店、定点单体药店、定点医疗机构(如定点医院)中各种类型的刷卡设备,可以是专门的刷卡机、也可以是安装有相应收款应用程序的各种智能移动设备,如手机、平板电脑等,以及各种图形采集设备,供医护人员下单、开具药房的终端设备的通信连通,从而可以与上述设备进行数据及指令的交互。
此外,定点医药机构监管平台还需要与社会保障平台,即现有的社保管理中心预先通信连通,这样就可以方便的获知各参保人员的信息。
此外,上述所说的第一生物特征信息,具体可以是人脸特征、指纹特征信息、虹膜特征信息、声纹特征信息等,此处不再一一列举,对此也不做任何限制,本领域的技术人员可以根据需要设置需要获取的第一生物特征信息。
相应地,在第一生物特征信息不同时,选取的获取第一生物特征信息的设备也会有所不同,比如在第一生物特征信息为人脸特征信息、虹膜特征信息时,则需要采用图像采集设备来获取;在第一生物特征信息为指纹特征信息时,则需要采用指纹模组获取;在第一生物特征信息为声纹特征信息时,需要采用语音装置来获取。
此外,上述所说的医疗保险卡,即社会医疗保险卡,也称医保卡。目前可以是集社保与医保于一体的社保卡。相应地,上述识别号具体是标识医疗保险卡唯一性的ID(Identification)号,或者是参保人的身份证号码。
为了便于理解获取使用医疗保险卡的用户的第一生物特征信息的具体实现方式,以下以第一生物特征信息为使用医疗保险卡的用户的第一人脸特征信息,第二生物特征信息为医疗保险卡的持有者的第二人脸特征信息为例进行具体说明。
具体的,在第一生物特征信息为使用医疗保险卡的用户的第一人脸特征信息时,获取使用医疗保险卡的用户的第一生物特征信息,包括:(1)接收定点医药机构中图像采集设备上传的包含使用医疗保险卡的用户的第一人脸图像的实时流媒体文件;(2)根据预存的人脸检测模型,对实时流媒体文件中的每一帧流媒体数据进行人脸检测,得到第一人脸图像;(3)根据预存的人脸特征提取模型,对第一人脸图像进行人脸特征提取,得到第一人脸特征信息。
需要说明的是,在具体实现中,上述人脸检测模型具体可以基于卷积神经网络算法对人脸样本数据进行卷积神经网络训练获得,人脸特征提取模型则可以根据人脸样本数据中的人脸特征训练获得。
关于卷积神经网络,本领域的技术人员可以知晓,其在机器学习中,是一种深度前馈人工神经网络,能够较为准确的识别图像中的信息。因此,通过对人脸样本数据(如预先从网络下载,或者拍摄录入的人脸图像)进行卷积神经网络训练,从而得到一个能够准确识别出图像中的人脸检测模型。
另外,值得一提的是,在具体应用中,采用的人脸检测模型具体是采用Caffe(卷积神经网络框架,Convolutional
Architecture for Fast Feature
Embedding)搭建,然后基于卷积神经网络算法训练获得。由于Caffe是一个清晰、可读性高、快速的深度学习框架,因此,基于Caffe来训练获得人脸检测模型,能够大大提高运行速度,并且能够大大减小训练所得的人脸检测模型的大小,使得训练所得的人脸检测模型为实时目标检测模,即无论图像采集设备上传的为静态图片还是实时流媒体文件,基于该人脸检测模型都可以从中提取出使用医疗保险卡的用户的第一人脸图像。
另外,为了提升人脸检测速度,人脸检测模型还可以采用现有人脸数据较为丰富的人脸识别数据库对实时快速目标检测(YOLO)神经网络模型进行卷积神经网络训练得到。
进一步的,为了保证训练获得的人脸检测模型能够更加精准,可以采用YOLO的升级版本YOLOv2神经网络模型进行卷积神经网络训练得到。
为了便于理解人脸检测模型的训练,以下进行举例说明:
编写脚本将人脸样本数据集转成YOLO可识别格式,配置YOLO的三个主要文件myobj.data(用于存储转换为YOLO可识别格式的人脸样本数据)、myobj.name(用于存储每个人脸样本数据对应的名称)、myobj.cfg(用于存储训练过程中所需的相关参数),在YOLO官网下载权值文件然后运行命令开始训练,观察avg(每次训练结果的平均值)这个值的变化,如果这个值基本不再变小,那么训练就可以停止了,此时就得到一个可以进行人脸检测的人脸检测模型。
而关于人脸特征提取模型的构建,在本实例中为了增加了训练模型的网络深度,使得训练出的人脸特征提取模型的提取精度能够大大提高,本实施例对现有的训练方式做了一下改进,具体为在进行训练之前,先将训练模型中的卷积核进行拆分,为了便于理解,以下对人脸特征提取模型构建模块所执行的操作进行具体描述:
首先,根据人脸样本数据中的人脸特征构建训练模型。
具体的说,此处所说的样本数据可以是利用网络预先从各大数据平台下载的人脸图像,也可以是预先拍摄并录入的人脸图像,此次不再一一列举,对此也不做任何限制。
然后,将训练模型中尺寸为的卷积核拆分为至少两个尺寸为的卷积核。
具体的拆分方式可以如图3和图4所示。
具体的说,图3中的字母A表示的为训练模型中尺寸为的卷积核,字母A1、A2为拆分后的四个尺寸为的卷积核;图4中的字母B表示的为训练模型中尺寸为的卷积核,字母B1、B2、B3和B4为拆分后的四个尺寸为的卷积核。
需要说明的是,以上给出的仅为两种具体的拆分方式,对本申请的技术方案并不构成任何限定,本领域的技术人员可以根据需要进行拆分,此处不做限制。
最后,基于卷积神经网络算法,对训练模型进行训练,得到人脸特征提取模型。
需要说明的是,由于本实施例中的人脸特征提取模型,也是一种卷积神经网络模型,即该人脸特征提取模型主要由卷积层、池化层和全连接层构成,其中卷积层和池化层的组合可以出现多次,全连接层位于池化层后,作为整个模型的输出层。
另外,本实施例中所说的人脸特征,具体采用人脸特征提取模型中输出层中节点的输出作为人脸特征,该人脸特征可以是由脸部的各个特征点构成的,如眼睛、鼻尖、嘴角点、眉毛以及脸部其他部件的轮廓点。
另外,需要说明的是,在实际应用中,全连接层可以有两个,如果全连接层为两个,则输出层为第二全连接层,具体的本领域的技术人员可以根据需要设置,此处不做限制。
另外,值得一提的是,为了加速后续训练过程中,人脸特征提取模型的收敛速度,并且在一定程度上提升人脸特征提取模型的泛化能力(机器学习算法对新鲜样本的适应能力),本实施方式中使用的人脸特征提取模型,在根据人脸样本数据中的人脸特征训练进行训练之前,可以先对人脸样本数据进行归一化处理,使得经过归一化处理后的人脸样本数据,能够把各个特征的尺度控制在相同的范围内,从而大大缩小了训练过程中每层卷积层中卷积核以及作为输出层的全连接层中的节点数,在简化训练过程中各种计算的同时,也可以提升构建的人脸特征提取模型的准确性。
步骤S20:根据识别号,从社会保障平台获取识别号对应的医疗保险卡的持有者的第二生物特征信息。
具体的说,上述所说的医疗保险卡的持有者具体是指该医疗保险卡对应的实际参保人。因此,根据识别号,从社会保障平台获取识别号对应的医疗保险卡的持有者的第二生物特征信息,即为参保人在办理医疗保险卡时,录入的人脸特征、指纹特征信息、虹膜特征信息、声纹特征信息等。
步骤S30:将第一生物特征信息与第二生物特征信息进行匹配。
仍以第一生物特征信息为第一人脸特征信息,第二生物特征信息为第二人脸特征信息为例,则在具体实现中将第一人脸特征信息与第二人脸特征信息进行匹配的具体操作大致如下:首先,将第一人脸特征信息与第二人脸特征信息进行逐一匹配,确定第一人脸特征信息与第二人脸特征信息之间的余弦相似度;然后,将余弦相似度与预设的相似度阈值进行比较。
步骤S40:若第一生物特征信息与第二生物特征信息不匹配,则向刷卡设备下发禁止使用医疗保险卡支付的指令,以使刷卡设备无法从医疗保险卡中扣费。
仍以第一生物特征信息为第一人脸特征信息,第二生物特征信息为第二人脸特征信息为例,则在将第一人脸特征信息与第二人脸特征信息进行匹配之后,若余弦相似度小于相似度阈值,则向刷卡设备下发禁止使用医疗保险卡支付的指令。
进一步地,为了方便监管部门对医保结算的核查,在执行上述步骤S10至步骤S40之前,还需要先对定点医药机构中采购的药品进行入库记录。
具体的,在定点医药机构采购了药品,并将药品入库时,需要将每一种的信息进行录入,并录入对应的数量、采购发票,相关部门规定价格等,关于录入的内容可以尽可能的详细,从而便于后期的监管和查询。
此外,为了给用户提供更好的购药就诊体验,可以在用户使用医疗保险卡购买药品时,由定点医疗机构的医护人员利用终端设备先查看医药库中是否还有用户选择的药品,如果没有,或者数量已经低于某一阈值,则及时作出提醒,以便根据用户的情况推荐其他合适的替代药品。如果有,则在支付过程中,按流程执行上述步骤S10至步骤S40。
同时,在本次交易完成后,为了保证后续正常使用,需要及时更新和维护药品的明细表,以记录各药品的剩余数量,同时记录该医保药品的出库时间,购买者的医疗保险卡的卡号、使用医疗保险卡支付的费用等。
进一步地,为了避免出现“无医师开方、医师不在岗开方”的问题,规范用药,避免由于处方不合理,用户随意购药导致的医保基金结算不合理增长。定点医药机构监管平台还可以预先获取各定点医疗机构的医疗人员的信息,并核查医疗人员是否具有行医资格,可以开具处方,从而避免定点医疗机构在不具备可以开具处方的医师时,随意开具处方。
此外,在判定各定点医疗机构具备有开处方的医师时,为了防止出现医师不在岗,其他人员随意开方,还可以设置对医师上岗时间的监管,比如采用各种防打卡设备,使得医师必需亲打卡,从而避免了在医师不在岗时,他人随意开处方。
应当理解的是,上述各步骤中列举的实例,以及给出的具体实现方式,对本申请的技术方案并不构成限定。在实际应用中,本领域的技术人员可以根据需要进行设置,此处不做限制。
通过上述描述不难发现,本实施例中提供的基于生物特征识别的定点医药机构的监管方法,在有用户使用医疗保险卡购买药品时,通过获取当前使用医疗保险卡购买药品的用户的第一生物特征和该医疗保险卡的识别号,并根据获取到的识别号从社会保障平台获取与当前识别号对应的医疗保险卡的真正的持有者的第二生物特征信息,然后将第一生物特征信息与第二生物特征信息进行匹配,在二者匹配时才能利用该医疗保险卡支付购买的药品的费用,否则便向发起刷卡请求的刷卡设备下发禁止使用该医疗保险卡支付的指令,使得刷卡设备无法从该医疗保险卡中成功扣费,从而可以有效的避免医疗定点医药机构利用监管漏洞进行违规刷卡。
参考图5,图5为本申请一种基于生物特征识别的定点医药机构的监管方法第二实施例的流程示意图。基于上述第一实施例,本实施例基于生物特征识别的定点医药机构的监管方法在步骤S30之后,还包括:
步骤S50:确定药品可以使用医疗保险卡购买,向刷卡设备下发允许使用医疗保险卡支付的指令,以使刷卡设备从医疗保险卡中扣费。
为了便于理解上述步骤S50的实现过程,以下进行具体说明:
具体的说,若通过执行步骤S30,确定第一生物特征信息与第二生物特征信息匹配,首先获取使用医疗保险卡的用户购买的药品的信息。
需要说明的是,此处所说的药品的信息,具体可以包括药品的名称、编号、生产厂家、药品类别、项目规格(如每盒多少袋,每袋多少克等)等,此处不再一一列举,对此也不做任何限制,本领域的技术人员可以根据需要设置需要获取的内容。
然后,根据预存的基本医疗保险药品目录和药品的信息,判断药品是否可以使用医疗保险卡购买,若药品可以使用医疗保险卡购买,则向刷卡设备下发允许使用医疗保险卡支付的指令,以使刷卡设备从医疗保险卡中扣费。
应当理解的是,上述所说的基本医疗保险药品目录具体可以直接选用政府医疗保险管理有关部门制定的“国家基本医疗保险药品目录”,即不做任何修改,也可以适应性的增加部分商业医疗保险支出报销的药品。
并且,为了方便查询,可以将上述基本医疗保险药品目录以表1的形式进行存储。
表1 基本医疗保险药品目录的存储表
编号 | 药品名称 | 发票名称 | 生产厂家 | 药品类别 | 项目规格 | 单位 | 是否医保 | 自付比例 |
20102049073321910101 | 复方感冒灵颗粒 | 成药费 | 广西宝瑞坦制药有限公司 | 乙类 | 14克/14袋 | 盒 | 是 | 无 |
20102048109197410501 | 复方感冒灵片 | 成药费 | 广东省罗浮山白鹤制药厂 | 乙类 | 6.25克/42毫克/100片 | 瓶 | 是 | 无 |
20102081234129510101 | 感冒清热胶囊 | 成药费 | 吉林敖东集团力源制药股份有限公司 | 乙类 | 0.45克/24粒 | 盒 | 是 | 无 |
需要说明的是,以上仅为举例说明,对本申请的技术方案并不构成任何限定,在具体实现中,本领域的技术人员可以根据需要构建基本医疗保险药品目录的存储表,此处不做限制。
通过上述描述不难发现,本实施例中提供的基于生物特征识别的定点医药机构的监管方法,通过将第一生物特征信息与第二生物特征信息进行匹配,并在第一生物特征信息与第二生物特征信息匹配时,获取使用医疗保险卡的用户购买的药品的信息,并根据预存的基本医疗保险药品目录和药品的信息,来判断药品是否可以使用医疗保险卡购买,在判定药品可以使用医疗保险卡购买时,才向刷卡设备下发允许使用医疗保险卡支付的指令,从而可以进一步避免定点医疗机构采用非医保药品代替医保药品,骗取医疗保险金,不仅给参保人带来损失,还导致了医疗保险基金的不合理使用。
此外,值得一提的是,在具体实现中,为了进一步杜绝定点医药机构为了自身利益,利用监管漏洞随意抬高药价,以次充好等现象的发生,在确定药品可以使用医疗保险卡购买,向刷卡设备下发允许使用医疗保险卡支付的指令之前,还可以进一步获取可以使用医疗保险卡购买的药品对应的购药订单,并从购药订单中提取药品的价格,然后根据预存的医保药品费用标准和药品的价格,判断药品价格是否满足医保药品费用标准的规定,若药品可以使用医疗保险卡购买,且药品价格满足医保药品费用标准的规定,则向刷卡设备下发允许使用医疗保险卡的指令。通过这种方式,可以实现对定点医药机构的有效监管,使得医疗保险基金能够被合理使用,保障了市场秩序,进而更好的保障了参保人员利益。
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,也可以通过计算机可读指令来指令相关的硬件完成,所述计算机可读指令可以存储于一种计算机可读存储介质中,上述提到的计算机可读存储介质可以是只读存储器,磁盘或光盘等。此外,本发明实施例还提出一种计算机可读存储介质,所述计算机可读存储介质上存储有计算机可读指令,所述计算机可读指令被处理器执行时实现如上文所述的基于生物特征识别的定点医药机构的监管方法的步骤。所述计算机可读存储介质可以是非易失性计算机可读存储介质。
参照图6,图6为本申请基于生物特征识别的定点医药机构的监管装置第一实施例的结构框图。如图6所示,本申请实施例提出的基于生物特征识别的定点医药机构的监管装置包括:第一获取模块6001、第二获取模块6002、匹配模块6003和发送模块6004。其中,第一获取模块6001,用于响应于定点医药机构中刷卡设备触发的使用医疗保险卡支付的刷卡请求,获取使用医疗保险卡的用户的第一生物特征信息和医疗保险卡的识别号;第二获取模块6002,用于根据识别号,从社会保障平台获取识别号对应的医疗保险卡的持有者的第二生物特征信息;匹配模块6003,用于将第一生物特征信息与第二生物特征信息进行匹配;发送模块6004,用于在第一生物特征信息与第二生物特征信息不匹配时,向刷卡设备下发禁止使用医疗保险卡支付的指令,以使刷卡设备无法从医疗保险卡中扣费。
此外,值得一提的是,在具体实现中第一获取模块6001获取到的第一生物特征信息具体可以为使用医疗保险卡的用户的第一人脸特征信息,第二获取模块6002获取到的第二生物特征信息具体可以为医疗保险卡的持有者的第二人脸特征信息。因而上述第一获取模块6001具体可以细化为实时流媒体文件接收子模块、人脸检测子模块和人脸特征信息提取子模块。
相应地,在第一生物特征信息为第一人脸特征信息时,第一获取模块6001获取使用医疗保险卡的用户的第一生物特征信息的操作具体如下:首先,由流媒体文件接收子模块,接收定点医药机构中图像采集设备上传的包含使用医疗保险卡的用户的第一人脸图像的实时流媒体文件;然后,由人脸检测子模块,根据预存的人脸检测模型,对实时流媒体文件中的每一帧流媒体数据进行人脸检测,得到第一人脸图像;最后,由人脸特征提取子模块,根据预存的人脸特征提取模型,对第一人脸图像进行人脸特征提取,得到第一人脸特征信息。
相应地,在第一生物特征信息为第一人脸特征信息时,匹配模块6003将第一生物特征信息与第二生物特征信息进行匹配的操作,具体为:将第一人脸特征信息与第二人脸特征信息进行匹配。
需要说明的是,在具体实现中,上述人脸检测模型具体可以基于卷积神经网络算法对人脸样本数据进行卷积神经网络训练获得,人脸特征提取模型则可以根据人脸样本数据中的人脸特征训练获得。
进一步地,为了保证第一获取模块6001获取使用医疗保险卡的用户的第一生物特征信息的操作能够顺利进行,需要预先构建上述用到的人脸检测模型和人脸特征提取模型。因此,在基于生物特征识别的定点医药机构的监管装置中还可以包括人脸检测模型构建模块和人脸特征提取模型构建模块。
关于卷积神经网络,本领域的技术人员可以知晓,其在机器学习中,是一种深度前馈人工神经网络,能够较为准确的识别图像中的信息。因此,通过对人脸样本数据(如预先从网络下载,或者拍摄录入的人脸图像)进行卷积神经网络训练,从而得到一个能够准确识别出图像中的人脸检测模型。关于其具体的训练流程,本领域的技术人员可以通过查找相关资料实现,此处不再赘述。
关于人脸特征提取模型的构建,本实施例对现有的训练方式做了一下改进,具体为在进行训练之前,先将训练模型中的卷积核进行拆分,为了便于理解,以下对人脸特征提取模型构建模块所执行的操作进行具体描述:首先,根据人脸样本数据中的人脸特征构建训练模型;然后,将训练模型中尺寸为的卷积核拆分为至少两个尺寸为的卷积核;最后,基于卷积神经网络算法,对训练模型进行训练,得到人脸特征提取模型。
需要说明的是,以上给出的仅为两种具体的拆分方式,对本申请的技术方案并不构成任何限定,本领域的技术人员可以根据需要进行拆分,此处不做限制。
此外,为了让不同维度之间的特征在数值上有一定的比较性,从而大大提高后续训练获得的人脸特征提取模型的准确性,本实例中提供的基于生物特征识别的定点医药机构的监管装置还可以包括归一化处理模块,从而可以在人脸特征提取模型构建模块根据人脸样本数据中的人脸特征构建训练模型之前,由归一化处理模块先对人脸样本数据进行归一化处理,使得经过归一化处理后的人脸样本数据,能够把各个特征的尺度控制在相同的范围内,进而提升构建的人脸特征提取模型的准确性。
通过上述描述不难发现,本实施例中,在有用户使用医疗保险卡购买药品时,通过获取当前使用医疗保险卡购买药品的用户的第一生物特征和该医疗保险卡的识别号,并根据获取到的识别号从社会保障平台获取与当前识别号对应的医疗保险卡的真正的持有者的第二生物特征信息,然后将第一生物特征信息与第二生物特征信息进行匹配,在二者匹配时才能利用该医疗保险卡支付购买的药品的费用,否则便向发起刷卡请求的刷卡设备下发禁止使用该医疗保险卡支付的指令,使得刷卡设备无法从该医疗保险卡中成功扣费,从而可以有效的避免医疗定点医药机构利用监管漏洞进行违规刷卡。
另外,未在本实施例中详尽描述的技术细节,可参见本申请任意实施例所提供的基于生物特征识别的定点医药机构的监管方法,此处不再赘述。
基于上述基于生物特征识别的定点医药机构的监管装置的第一实施例,提出本申请基于生物特征识别的定点医药机构的监管装置第二实施例。
在本实施例中,基于生物特征识别的定点医药机构的监管装置还包括药品信息获取模块、医保药品判断模块和第二发送模块。其中,药品信息获取模块,用于在第一生物特征信息与第二生物特征信息匹配时,获取使用医疗保险卡的用户购买的药品的信息;医保药品判断模块,用于根据预存的基本医疗保险药品目录和药品的信息,判断药品是否可以使用医疗保险卡购买;第二发送模块,用于在药品可以使用医疗保险卡购买时,向刷卡设备下发允许使用医疗保险卡支付的指令,以使刷卡设备从医疗保险卡中扣费。
通过上述描述不难发现,本实施例中,通过将第一生物特征信息与第二生物特征信息进行匹配,并在第一生物特征信息与第二生物特征信息匹配时,获取使用医疗保险卡的用户购买的药品的信息,并根据预存的基本医疗保险药品目录和药品的信息,来判断药品是否可以使用医疗保险卡购买,在判定药品可以使用医疗保险卡购买时,才向刷卡设备下发允许使用医疗保险卡支付的指令,从而可以进一步避免定点医疗机构采用非医保药品代替医保药品,骗取医疗保险金,不仅给参保人带来损失,还导致了医疗保险基金的不合理使用。
此外,值得一提的是,在具体实现中,为了进一步杜绝定点医药机构为了自身利益,利用监管漏洞随意抬高药价,以次充好等现象的发生,本实施例中提供的订单医药机构的监管装置还可以包括:医保药品价格合理性判断模块,从而可以在医保药品判断模块确定药品可以使用医疗保险卡购买,向刷卡设备下发允许使用医疗保险卡支付的指令之前,通过医保药品价格合理性判断模块获取可以使用医疗保险卡购买的药品对应的购药订单,并从购药订单中提取药品的价格,然后根据预存的医保药品费用标准和药品的价格,判断药品价格是否满足医保药品费用标准的规定,从而可以控制向刷卡设备下发允许使用医疗保险卡的指令的第二发送模块,在药品可以使用医疗保险卡购买,且药品价格满足医保药品费用标准的规定时,才向刷卡设备下发允许使用医疗保险卡的指令。通过这种方式,可以实现对定点医药机构的有效监管,使得医疗保险基金能够被合理使用,保障了市场秩序,进而更好的保障了参保人员利益。
另外,未在本实施例中详尽描述的技术细节,可参见本申请任意实施例所提供的基于生物特征识别的定点医药机构的监管方法,此处不再赘述。
此外,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一个计算机可读存储介质(如只读存储器(Read Only
Memory,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)
- 一种基于生物特征识别的定点医药机构的监管方法,其特征在于,所述方法包括以下步骤:定点医药机构监管平台响应于定点医药机构中刷卡设备触发的使用医疗保险卡支付的刷卡请求,获取使用所述医疗保险卡的用户的第一生物特征信息和所述医疗保险卡的识别号;根据所述识别号,从社会保障平台获取所述识别号对应的所述医疗保险卡的持有者的第二生物特征信息;将所述第一生物特征信息与所述第二生物特征信息进行匹配;若所述第一生物特征信息与所述第二生物特征信息不匹配,则向所述刷卡设备下发禁止使用所述医疗保险卡支付的指令,以使所述刷卡设备无法从所述医疗保险卡中扣费。
- 如权利要求1所述的方法,其特征在于,所述第一生物特征信息为使用所述医疗保险卡的用户的第一人脸特征信息,所述第二生物特征信息为所述医疗保险卡的持有者的第二人脸特征信息;所述获取使用所述医疗保险卡的用户的第一生物特征信息的步骤,包括:接收所述定点医药机构中图像采集设备上传的包含使用所述医疗保险卡的用户的第一人脸图像的实时流媒体文件;根据预存的人脸检测模型,对所述实时流媒体文件中的每一帧流媒体数据进行人脸检测,得到所述第一人脸图像,所述人脸检测模型基于卷积神经网络算法对人脸样本数据进行卷积神经网络训练获得;根据预存的人脸特征提取模型,对所述第一人脸图像进行人脸特征提取,得到所述第一人脸特征信息,所述人脸特征提取模型根据所述人脸样本数据中的人脸特征训练获得;其中,所述将所述第一生物特征信息与所述第二生物特征信息进行匹配,包括:将所述第一人脸特征信息与所述第二人脸特征信息进行匹配。
- 如权利要求2所述的方法,其特征在于,所述将所述第一人脸特征信息与所述第二人脸特征信息进行匹配的步骤,包括:将所述第一人脸特征信息与所述第二人脸特征信息进行逐一匹配,确定所述第一人脸特征信息与所述第二人脸特征信息之间的余弦相似度;将所述余弦相似度与预设的相似度阈值进行比较;其中,所述若所述第一生物特征信息与所述第二生物特征信息不匹配,则向所述刷卡设备下发禁止使用所述医疗保险卡支付的指令,包括:若所述余弦相似度小于所述相似度阈值,则向所述刷卡设备下发禁止使用所述医疗保险卡支付的指令。
- 如权利要求2或3所述的方法,其特征在于,所述获取使用所述医疗保险卡的用户的第一生物特征信息的步骤之前,所述方法还包括以下步骤:构建所述人脸特征提取模型;所述构建所述人脸特征提取模型,包括:根据所述人脸样本数据中的人脸特征构建训练模型;将所述训练模型中尺寸为的卷积核拆分为至少两个尺寸为的卷积核;基于所述卷积神经网络算法,对所述训练模型进行训练,得到所述人脸特征提取模型。
- 如权利要求4所述的方法,其特征在于,所述根据所述人脸样本数据中的人脸特征构建训练模型的步骤之前,所述方法还包括以下步骤:对所述人脸样本数据进行归一化处理。
- 如权利要求1所述的方法,其特征在于,所述将所述第一生物特征信息与所述第二生物特征信息进行匹配的步骤之后,所述方法还包括以下步骤:若所述第一生物特征信息与所述第二生物特征信息匹配,则获取使用所述医疗保险卡的用户购买的药品的信息;根据预存的基本医疗保险药品目录和所述药品的信息,判断所述药品是否可以使用所述医疗保险卡购买;若所述药品可以使用所述医疗保险卡购买,则向所述刷卡设备下发允许使用所述医疗保险卡支付的指令,以使所述刷卡设备从所述医疗保险卡中扣费。
- 如权利要求6所述的方法,其特征在于,所述若所述药品可以使用所述医疗保险卡购买,则向所述刷卡设备下发允许使用所述医疗保险卡支付的指令的步骤之前,所述方法还包括以下步骤:获取可以使用所述医疗保险卡购买的药品对应的购药订单,从所述购药订单中提取所述药品的价格;根据预存的医保药品费用标准和所述药品的价格,判断所述药品价格是否满足所述医保药品费用标准的规定;其中,所述若所述药品可以使用所述医疗保险卡购买,则向所述刷卡设备下发允许使用所述医疗保险卡的指令,包括:若所述药品可以使用所述医疗保险卡购买,且所述药品价格满足所述医保药品费用标准的规定,则向所述刷卡设备下发允许使用所述医疗保险卡的指令。
- 一种基于生物特征识别的定点医药机构的监管装置,其特征在于,所述装置包括:第一获取模块,用于响应于定点医药机构中刷卡设备触发的使用医疗保险卡支付的刷卡请求,获取使用所述医疗保险卡的用户的第一生物特征信息和所述医疗保险卡的识别号;第二获取模块,用于根据所述识别号,从社会保障平台获取所述识别号对应的所述医疗保险卡的持有者的第二生物特征信息;匹配模块,用于将所述第一生物特征信息与所述第二生物特征信息进行匹配;发送模块,用于在所述第一生物特征信息与所述第二生物特征信息不匹配时,向所述刷卡设备下发禁止使用所述医疗保险卡支付的指令,以使所述刷卡设备无法从所述医疗保险卡中扣费。
- 一种基于生物特征识别的定点医药机构的监管设备,其特征在于,所述设备包括:存储器、处理器及存储在所述存储器上并可在所述处理器上运行的计算机可读存储指令,所述计算机可读指令被处理器执行时,使得所述处理器执行以下步骤:定点医药机构监管平台响应于定点医药机构中刷卡设备触发的使用医疗保险卡支付的刷卡请求,获取使用所述医疗保险卡的用户的第一生物特征信息和所述医疗保险卡的识别号;根据所述识别号,从社会保障平台获取所述识别号对应的所述医疗保险卡的持有者的第二生物特征信息;将所述第一生物特征信息与所述第二生物特征信息进行匹配;若所述第一生物特征信息与所述第二生物特征信息不匹配,则向所述刷卡设备下发禁止使用所述医疗保险卡支付的指令,以使所述刷卡设备无法从所述医疗保险卡中扣费。
- 如权利要求9所述的设备,其特征在于,所述第一生物特征信息为使用所述医疗保险卡的用户的第一人脸特征信息,所述第二生物特征信息为所述医疗保险卡的持有者的第二人脸特征信息;所述获取使用所述医疗保险卡的用户的第一生物特征信息的步骤,包括:接收所述定点医药机构中图像采集设备上传的包含使用所述医疗保险卡的用户的第一人脸图像的实时流媒体文件;根据预存的人脸检测模型,对所述实时流媒体文件中的每一帧流媒体数据进行人脸检测,得到所述第一人脸图像,所述人脸检测模型基于卷积神经网络算法对人脸样本数据进行卷积神经网络训练获得;根据预存的人脸特征提取模型,对所述第一人脸图像进行人脸特征提取,得到所述第一人脸特征信息,所述人脸特征提取模型根据所述人脸样本数据中的人脸特征训练获得;其中,所述将所述第一生物特征信息与所述第二生物特征信息进行匹配,包括:将所述第一人脸特征信息与所述第二人脸特征信息进行匹配。
- 如权利要求10所述的设备,其特征在于,所述将所述第一人脸特征信息与所述第二人脸特征信息进行匹配的步骤,包括:将所述第一人脸特征信息与所述第二人脸特征信息进行逐一匹配,确定所述第一人脸特征信息与所述第二人脸特征信息之间的余弦相似度;将所述余弦相似度与预设的相似度阈值进行比较;其中,所述若所述第一生物特征信息与所述第二生物特征信息不匹配,则向所述刷卡设备下发禁止使用所述医疗保险卡支付的指令,包括:若所述余弦相似度小于所述相似度阈值,则向所述刷卡设备下发禁止使用所述医疗保险卡支付的指令。
- 如权利要求10或11所述的设备,其特征在于,所述获取使用所述医疗保险卡的用户的第一生物特征信息的步骤之前,所述处理器还用于执行以下步骤:构建所述人脸特征提取模型;所述构建所述人脸特征提取模型,包括:根据所述人脸样本数据中的人脸特征构建训练模型;将所述训练模型中尺寸为的卷积核拆分为至少两个尺寸为的卷积核;基于所述卷积神经网络算法,对所述训练模型进行训练,得到所述人脸特征提取模型。
- 如权利要求12所述的设备,其特征在于,所述根据所述人脸样本数据中的人脸特征构建训练模型的步骤之前,所述处理器还用于执行以下步骤:对所述人脸样本数据进行归一化处理。
- 如权利要求9所述的设备,其特征在于,所述将所述第一生物特征信息与所述第二生物特征信息进行匹配的步骤之后,所述处理器还用于执行以下步骤:若所述第一生物特征信息与所述第二生物特征信息匹配,则获取使用所述医疗保险卡的用户购买的药品的信息;根据预存的基本医疗保险药品目录和所述药品的信息,判断所述药品是否可以使用所述医疗保险卡购买;若所述药品可以使用所述医疗保险卡购买,则向所述刷卡设备下发允许使用所述医疗保险卡支付的指令,以使所述刷卡设备从所述医疗保险卡中扣费。
- 如权利要求14所述的设备,其特征在于,所述若所述药品可以使用所述医疗保险卡购买,则向所述刷卡设备下发允许使用所述医疗保险卡支付的指令的步骤之前,所述处理器还用于执行以下步骤:获取可以使用所述医疗保险卡购买的药品对应的购药订单,从所述购药订单中提取所述药品的价格;根据预存的医保药品费用标准和所述药品的价格,判断所述药品价格是否满足所述医保药品费用标准的规定;其中,所述若所述药品可以使用所述医疗保险卡购买,则向所述刷卡设备下发允许使用所述医疗保险卡的指令,包括:若所述药品可以使用所述医疗保险卡购买,且所述药品价格满足所述医保药品费用标准的规定,则向所述刷卡设备下发允许使用所述医疗保险卡的指令。
- 一种计算机可读存储介质,其特征在于,所述计算机可读存储介质上存储有计算机可读指令,所述计算机可读指令被处理器执行时实现以下步骤:定点医药机构监管平台响应于定点医药机构中刷卡设备触发的使用医疗保险卡支付的刷卡请求,获取使用所述医疗保险卡的用户的第一生物特征信息和所述医疗保险卡的识别号;根据所述识别号,从社会保障平台获取所述识别号对应的所述医疗保险卡的持有者的第二生物特征信息;将所述第一生物特征信息与所述第二生物特征信息进行匹配;若所述第一生物特征信息与所述第二生物特征信息不匹配,则向所述刷卡设备下发禁止使用所述医疗保险卡支付的指令,以使所述刷卡设备无法从所述医疗保险卡中扣费。
- 如权利要求16所述的计算机可读存储介质,其特征在于,所述第一生物特征信息为使用所述医疗保险卡的用户的第一人脸特征信息,所述第二生物特征信息为所述医疗保险卡的持有者的第二人脸特征信息;所述获取使用所述医疗保险卡的用户的第一生物特征信息的步骤,包括:接收所述定点医药机构中图像采集设备上传的包含使用所述医疗保险卡的用户的第一人脸图像的实时流媒体文件;根据预存的人脸检测模型,对所述实时流媒体文件中的每一帧流媒体数据进行人脸检测,得到所述第一人脸图像,所述人脸检测模型基于卷积神经网络算法对人脸样本数据进行卷积神经网络训练获得;根据预存的人脸特征提取模型,对所述第一人脸图像进行人脸特征提取,得到所述第一人脸特征信息,所述人脸特征提取模型根据所述人脸样本数据中的人脸特征训练获得;其中,所述将所述第一生物特征信息与所述第二生物特征信息进行匹配,包括:将所述第一人脸特征信息与所述第二人脸特征信息进行匹配。
- 如权利要求17所述的计算机可读存储介质,其特征在于,所述将所述第一人脸特征信息与所述第二人脸特征信息进行匹配的步骤,包括:将所述第一人脸特征信息与所述第二人脸特征信息进行逐一匹配,确定所述第一人脸特征信息与所述第二人脸特征信息之间的余弦相似度;将所述余弦相似度与预设的相似度阈值进行比较;其中,所述若所述第一生物特征信息与所述第二生物特征信息不匹配,则向所述刷卡设备下发禁止使用所述医疗保险卡支付的指令,包括:若所述余弦相似度小于所述相似度阈值,则向所述刷卡设备下发禁止使用所述医疗保险卡支付的指令。
- 如权利要求17或18所述的计算机可读存储介质,其特征在于,所述获取使用所述医疗保险卡的用户的第一生物特征信息的步骤之前,所述处理器还用于执行以下步骤:构建所述人脸特征提取模型;所述构建所述人脸特征提取模型,包括:根据所述人脸样本数据中的人脸特征构建训练模型;将所述训练模型中尺寸为的卷积核拆分为至少两个尺寸为的卷积核;基于所述卷积神经网络算法,对所述训练模型进行训练,得到所述人脸特征提取模型。
- 如权利要求19所述的计算机可读存储介质,其特征在于,所述根据所述人脸样本数据中的人脸特征构建训练模型的步骤之前,所述处理器还用于执行以下步骤:对所述人脸样本数据进行归一化处理。
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811473507.3A CN109598630A (zh) | 2018-12-04 | 2018-12-04 | 基于生物特征识别的定点医药机构的监管方法及相关设备 |
CN201811473507.3 | 2018-12-04 |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2020114221A1 true WO2020114221A1 (zh) | 2020-06-11 |
Family
ID=65960944
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/CN2019/118818 WO2020114221A1 (zh) | 2018-12-04 | 2019-11-15 | 基于生物特征识别的定点医药机构的监管方法及相关设备 |
Country Status (2)
Country | Link |
---|---|
CN (1) | CN109598630A (zh) |
WO (1) | WO2020114221A1 (zh) |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112925819A (zh) * | 2020-12-21 | 2021-06-08 | 上海药慧信息技术有限公司 | 药品中标信息挖掘方法和装置 |
CN113688784A (zh) * | 2021-09-10 | 2021-11-23 | 平安医疗健康管理股份有限公司 | 基于人脸识别的医保卡盗用风险识别方法及其相关设备 |
CN116525060A (zh) * | 2023-06-25 | 2023-08-01 | 中南大学 | 基于医药制品监管模型的智能区域监控方法及系统 |
CN118333533A (zh) * | 2024-06-17 | 2024-07-12 | 南京市医疗保障局 | 一种结合指标率的招采结算系统 |
Families Citing this family (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109598630A (zh) * | 2018-12-04 | 2019-04-09 | 平安医疗健康管理股份有限公司 | 基于生物特征识别的定点医药机构的监管方法及相关设备 |
CN111539833B (zh) * | 2020-04-10 | 2023-01-10 | 支付宝(杭州)信息技术有限公司 | 医疗费用支付方法、装置和系统 |
CN113159973A (zh) * | 2021-02-25 | 2021-07-23 | 华夏方圆信用评估有限公司 | 智慧医保基金动态监管方法及系统 |
CN113283894A (zh) * | 2021-06-07 | 2021-08-20 | 支付宝(杭州)信息技术有限公司 | 基于医保电子凭证进行支付的方法、装置和计算机可读介质 |
CN114037541A (zh) * | 2021-11-05 | 2022-02-11 | 湖南创研科技股份有限公司 | 基于生物特征识别的定点医药机构的监管方法及相关设备 |
CN114037540A (zh) * | 2021-11-05 | 2022-02-11 | 湖南创研科技股份有限公司 | 一种医保基金监管系统及其监控方法 |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107920070A (zh) * | 2017-11-15 | 2018-04-17 | 中移在线服务有限公司 | 身份认证方法、服务器和系统 |
CN108596814A (zh) * | 2018-04-20 | 2018-09-28 | 聊城农村商业银行股份有限公司 | 基于人脸识别的社保认证系统 |
CN108665559A (zh) * | 2018-05-21 | 2018-10-16 | 中兴智能视觉大数据技术(湖北)有限公司 | 一种人证票核实管理系统和方法 |
CN108876297A (zh) * | 2018-06-14 | 2018-11-23 | 平安健康保险股份有限公司 | 用于商业医保的自动审核方法、装置、设备及存储介质 |
CN109598630A (zh) * | 2018-12-04 | 2019-04-09 | 平安医疗健康管理股份有限公司 | 基于生物特征识别的定点医药机构的监管方法及相关设备 |
Family Cites Families (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105868977A (zh) * | 2016-04-01 | 2016-08-17 | 深圳市至高通信技术发展有限公司 | 药品售卖监管的方法、系统及装置 |
CN208077174U (zh) * | 2017-11-15 | 2018-11-09 | 中移在线服务有限公司 | 用于身份认证的识别装置 |
-
2018
- 2018-12-04 CN CN201811473507.3A patent/CN109598630A/zh active Pending
-
2019
- 2019-11-15 WO PCT/CN2019/118818 patent/WO2020114221A1/zh active Application Filing
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107920070A (zh) * | 2017-11-15 | 2018-04-17 | 中移在线服务有限公司 | 身份认证方法、服务器和系统 |
CN108596814A (zh) * | 2018-04-20 | 2018-09-28 | 聊城农村商业银行股份有限公司 | 基于人脸识别的社保认证系统 |
CN108665559A (zh) * | 2018-05-21 | 2018-10-16 | 中兴智能视觉大数据技术(湖北)有限公司 | 一种人证票核实管理系统和方法 |
CN108876297A (zh) * | 2018-06-14 | 2018-11-23 | 平安健康保险股份有限公司 | 用于商业医保的自动审核方法、装置、设备及存储介质 |
CN109598630A (zh) * | 2018-12-04 | 2019-04-09 | 平安医疗健康管理股份有限公司 | 基于生物特征识别的定点医药机构的监管方法及相关设备 |
Cited By (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112925819A (zh) * | 2020-12-21 | 2021-06-08 | 上海药慧信息技术有限公司 | 药品中标信息挖掘方法和装置 |
CN112925819B (zh) * | 2020-12-21 | 2023-05-19 | 上海药慧信息技术有限公司 | 药品中标信息挖掘方法和装置 |
CN113688784A (zh) * | 2021-09-10 | 2021-11-23 | 平安医疗健康管理股份有限公司 | 基于人脸识别的医保卡盗用风险识别方法及其相关设备 |
CN113688784B (zh) * | 2021-09-10 | 2024-05-14 | 平安医疗健康管理股份有限公司 | 基于人脸识别的医保卡盗用风险识别方法及其相关设备 |
CN116525060A (zh) * | 2023-06-25 | 2023-08-01 | 中南大学 | 基于医药制品监管模型的智能区域监控方法及系统 |
CN116525060B (zh) * | 2023-06-25 | 2023-10-03 | 中南大学 | 基于医药制品监管模型的智能区域监控方法及系统 |
CN118333533A (zh) * | 2024-06-17 | 2024-07-12 | 南京市医疗保障局 | 一种结合指标率的招采结算系统 |
Also Published As
Publication number | Publication date |
---|---|
CN109598630A (zh) | 2019-04-09 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
WO2020114221A1 (zh) | 基于生物特征识别的定点医药机构的监管方法及相关设备 | |
WO2020078053A1 (zh) | 医疗数据异常检测方法、装置、设备及存储介质 | |
US20190311377A1 (en) | Social security fraud behaviors identification method, device, apparatus and computer-readable storage medium | |
WO2020213843A1 (ko) | 사용자 맞춤형 의료정보 제공 시스템 및 이의 구동방법 | |
WO2020211388A1 (zh) | 基于预测模型的行为预测方法、装置、设备及存储介质 | |
WO2020024398A1 (zh) | 生物特征辅助支付方法、装置、计算机设备及存储介质 | |
US10897461B2 (en) | Pharmacy database access methods and systems | |
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 | |
WO2020078058A1 (zh) | 医疗数据异常识别方法、装置、终端及存储介质 | |
WO2016165177A1 (zh) | 基于面部识别的网络医院自助缴费方法和系统 | |
JP6897953B2 (ja) | 入場受付端末、入場受付方法、入場受付プログラム、および入場受付システム | |
KR20040099433A (ko) | 네트워크를 통해 백신 및 질병 데이터베이스와커뮤니케이션 되는 원격 스테이션을 갖는, 예방접종데이터를 수집, 저장, 표현, 및 분석하기 위한 시스템 | |
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 | |
US11386987B2 (en) | Providing global accessibility to telehealth prescribed medications | |
WO2020238915A1 (zh) | 用户身份识别方法及装置、电子设备、存储介质 | |
WO2020077837A1 (zh) | 业务数据核算方法、装置、设备及计算机可读存储介质 | |
US20170344709A1 (en) | Medical patient billing session using a temporary identity | |
CN108009435A (zh) | 数据脱敏方法、装置及存储介质 | |
WO2020119401A1 (zh) | 医师违规执业的监控方法、监控服务端及存储介质 | |
WO2019240343A1 (ko) | 사용자 상태에 따른 맞춤형 자기관리정보 제공시스템 | |
WO2016165181A1 (zh) | 基于指纹识别的网络医院自助缴费方法和系统 | |
WO2019088786A1 (ko) | 의약품 유효기한 알림 방법 및 이를 위한 컴퓨터 프로그램 기록매체 | |
US20230107711A1 (en) | Systems and methods for identifying a secondary card user | |
WO2016165180A1 (zh) | 基于网络医院挂号号码的自助缴费方法和系统 | |
WO2016133284A1 (ko) | 기부 매개 방법 |
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: 19893747 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: 19893747 Country of ref document: EP Kind code of ref document: A1 |
|
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) |