WO2023178978A1 - 基于人工智能的处方审核方法、装置、设备及介质 - Google Patents

基于人工智能的处方审核方法、装置、设备及介质 Download PDF

Info

Publication number
WO2023178978A1
WO2023178978A1 PCT/CN2022/122999 CN2022122999W WO2023178978A1 WO 2023178978 A1 WO2023178978 A1 WO 2023178978A1 CN 2022122999 W CN2022122999 W CN 2022122999W WO 2023178978 A1 WO2023178978 A1 WO 2023178978A1
Authority
WO
WIPO (PCT)
Prior art keywords
information
prescription
drug
user
drug information
Prior art date
Application number
PCT/CN2022/122999
Other languages
English (en)
French (fr)
Inventor
崔东超
王安宇
Original Assignee
康键信息技术(深圳)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 康键信息技术(深圳)有限公司 filed Critical 康键信息技术(深圳)有限公司
Publication of WO2023178978A1 publication Critical patent/WO2023178978A1/zh

Links

Classifications

    • GPHYSICS
    • G16INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16HHEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
    • G16H20/00ICT specially adapted for therapies or health-improving plans, e.g. for handling prescriptions, for steering therapy or for monitoring patient compliance
    • G16H20/10ICT specially adapted for therapies or health-improving plans, e.g. for handling prescriptions, for steering therapy or for monitoring patient compliance relating to drugs or medications, e.g. for ensuring correct administration to patients

Definitions

  • This application relates to the field of artificial intelligence technology, and in particular to a prescription review method, device, electronic equipment and computer-readable storage medium based on artificial intelligence.
  • This application provides an artificial intelligence-based prescription review method, including:
  • the prescription form is reviewed based on the drug information, the prohibited drug information and the applicable drug information, and the review results are output.
  • This application also provides an artificial intelligence-based prescription review device, which includes:
  • a prescription information judgment module used to receive user input information and judge whether there is prescription information in the input information
  • a prescription order generation module configured to generate a prescription order based on the prescription information when there is prescription information in the input information; to send a prescription information collection request to the user when there is no prescription information in the input information; to receive the user's request for collecting prescription information based on the prescription information. receive the prescription information returned by the prescription information collection request, and generate a prescription order based on the returned prescription information;
  • a prohibited drug information generation module is used to extract drug user information, condition information and drug information based on the prescription form, and perform medical history analysis based on the drug user information to obtain prohibited drug information;
  • Applicable drug information generation module used to construct a target vector matrix based on the condition information, and use a pre-built condition analysis model to calculate the target vector matrix to obtain applicable drug information;
  • a prescription order review module is used to review the prescription order based on the drug information, the prohibited drug information and the applicable drug information, and output the review results.
  • This application also provides an electronic device, which includes:
  • the memory stores a computer program executable by the at least one processor, the computer program being executed by the at least one processor to enable the at least one processor to perform artificial intelligence-based prescriptions as described below Review method:
  • the prescription form is reviewed based on the drug information, the prohibited drug information and the applicable drug information, and the review results are output.
  • the present application also provides a computer-readable storage medium in which at least one computer program is stored, and the at least one computer program is executed by a processor in an electronic device to implement artificial intelligence-based methods as described below.
  • the prescription form is reviewed based on the drug information, the prohibited drug information and the applicable drug information, and the review results are output.
  • Figure 1 is a schematic flow chart of an artificial intelligence-based prescription review method provided by an embodiment of the present application
  • Figure 2 is a schematic flow chart of generating a prescription order from returned prescription information provided by an embodiment of the present application
  • Figure 3 is a schematic flow chart of medical history analysis provided by an embodiment of the present application.
  • Figure 4 is a functional module diagram of an artificial intelligence-based prescription review device provided by an embodiment of the present application.
  • FIG. 5 is a schematic structural diagram of an electronic device that implements the artificial intelligence-based prescription review method provided by an embodiment of the present application.
  • the embodiment of this application provides a prescription review method based on artificial intelligence.
  • the execution subject of the artificial intelligence-based prescription review method includes, but is not limited to, at least one of electronic devices such as servers and terminals that can be configured to execute the method provided by the embodiments of the present application.
  • the artificial intelligence-based prescription review method can be executed by software or hardware installed on the terminal device or the server device, and the software can be a blockchain platform.
  • the server includes but is not limited to: a single server, a server cluster, a cloud server or a cloud server cluster, etc.
  • the server may be an independent server, or may provide cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, and content delivery networks (ContentDeliveryNetwork, CDN), as well as cloud servers for basic cloud computing services such as big data and artificial intelligence platforms.
  • cloud services such as cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, and content delivery networks (ContentDeliveryNetwork, CDN), as well as cloud servers for basic cloud computing services such as big data and artificial intelligence platforms.
  • the artificial intelligence-based prescription review method includes:
  • the user's input information may be pictures of prescription orders issued by hospitals or clinics, tables containing prescription information imported from other channels, etc.
  • determining whether there is prescription information in the input information includes:
  • the information format includes a preset format, it is determined that the input information contains prescription information
  • the information format does not include a preset format, it is determined that there is no prescription information in the input information.
  • the information format may be JPG, BMP, PNG, EXCEL, etc.; if the preset information format is JPG, BMP, PNG, EXCEL, and when the input information contains information formats such as JPG, BMP, PNG, etc., then the The above information format includes a preset format, that is, the input information contains a prescription order image, and the prescription order image is the prescription information.
  • the prescription information may be a display form containing prescription information such as a prescription order image.
  • generating a prescription order based on the prescription information includes:
  • a preset prescription template is used to generate a prescription.
  • performing text recognition on the prescription information to obtain the prescription text includes:
  • Optical character text recognition is performed on the text area to obtain prescription text.
  • the prescription information may be a prescription order image
  • the text type includes a printed text type and a handwritten text type
  • an image classification model may be used to perform text type recognition on the prescription information, and the image classification model may It is a deep learning model based on the ResNet18 algorithm
  • the image text type corresponding to the target text image can be the image text type corresponding to the text area to be recognized of the target text image, or it can also be the image text type corresponding to the entire target text image.
  • an optical character text recognition model can be used to perform optical character text recognition on the text area. If the text type is a handwritten text type, the text area to be recognized is input into the preset handwritten optical character text recognition model; if the text type is a printed text type, the text area to be recognized is input into the preset text recognition model. Designed printed optical character text recognition model. Both the handwritten optical character text recognition model and the printed optical character text recognition model can be convolutional recurrent neural network (CRNN) models.
  • CRNN convolutional recurrent neural network
  • a prescription information collection request needs to be sent to the user to remind the user to actively input.
  • the prescription information returned by the user according to the prescription information collection request may be a supplementary picture of a prescription issued by a hospital or clinic, a form containing prescription information imported from other channels, or the user may return a prescription information according to a prescription information collection request.
  • Personal medical records filled out in the attached template may be a supplementary picture of a prescription issued by a hospital or clinic, a form containing prescription information imported from other channels, or the user may return a prescription information according to a prescription information collection request.
  • the prescription form generated based on the returned prescription information includes:
  • a deep learning text classification model (such as fastText model, TextCNN, TextRNN, TextRNN + Attention, TextRCNN, etc.) in natural language learning (NPL) can be used to perform semantic analysis on the returned prescription information.
  • NPL natural language learning
  • cosine similarity Pearson correlation coefficient, Euclidean distance, etc. can be used to calculate the similarity between multiple semantic paragraphs and the template paragraph of the preset prescription template.
  • a prescription information collection request can be sent to the user with a prescription form filling template.
  • the user can directly input information according to the prescription form template.
  • the entered information is the prescription information, and the user returns the prescription information.
  • the prescription form can be generated directly from the prescription form template.
  • the drug user information may include real-name authenticated personal information, past medical history, drug allergy history, family genetic disease and other information.
  • the condition information may be a user's description of the condition, and the drug information may be The name of the specific medicine the user wants to purchase, etc.
  • the prescription order can be a document, form, etc. with a fixed format template.
  • the corresponding user information, condition information, or drug information can be obtained.
  • the prohibited drug information is drugs that are prohibited or should be taken with caution based on the past medical history, personal information, etc. of the drug user.
  • the medical history analysis is performed based on the drug user information to obtain banned drug information, including:
  • the banned drug information database contains different medical histories and corresponding banned drugs, banned drugs corresponding to different age groups, etc.
  • the banned drug information includes prednisone pine, dexamethasone, betamethasone; if there is an information segment "5 years old", this segment can be retrieved in the banned drug information database, and the corresponding banned drugs include hydroxychloroquine, imipramine, ranitidine, etc.
  • the prohibited drug information includes hydroxychloroquine, imipramine, and ranitidine.
  • the applicable drug information is the drug information that should be used based on the condition information.
  • constructing a target vector matrix based on the condition information includes:
  • Extract data to be masked from the condition information perform a masking operation on the data to be masked, and obtain masked data
  • the positioning vector set is converted into a positioning vector matrix, and the positioning vector matrix is used to adjust the iterative weight factor in the pre-constructed feedforward neural network to obtain a target vector matrix.
  • keywords can be extracted from the data to be masked according to the preset masking probability, and a masking operation is performed on the keywords to obtain masked words; in the data to be masked , replace the keyword with the masked word to obtain the masked data.
  • the Word2vec algorithm can be used to convert all the data in the masked data into vectors.
  • the method before using the pre-constructed condition analysis model to calculate the target vector matrix to obtain applicable drug information, the method further includes:
  • the preset disease name loss function is used to calculate the loss value between the output result and the real adapted drug information, and the condition analysis model is optimized according to the loss value to obtain a standard condition analysis model.
  • condition analysis model is a pre-trained language model, including but not limited to the BERT model (BidirectionalEncoderRepresentationsfromTransformers, bidirectional encoding representation model) and the LSTM model (Long-Short Term Memory, long short-term memory model).
  • BERT model BidirectionalEncoderRepresentationsfromTransformers, bidirectional encoding representation model
  • LSTM model Long-Short Term Memory, long short-term memory model
  • the target vector matrix is calculated using a pre-constructed condition analysis model to obtain applicable drug information, including:
  • the condition analysis model performs a preset number of convolutions, pooling and full connections on the target vector matrix to obtain condition analysis information
  • the applicable drug information corresponding to the condition analysis information is calculated through the activator.
  • the output result can be used as applicable drug information.
  • the prescription form is reviewed based on the drug information, the prohibited drug information and the applicable drug information, and the review results are output, including:
  • prohibited drug information and applicable drug information are obtained through analysis of drug user information and condition information respectively, and the scope of drugs has been divided. As long as information comparison is performed based on the drug information, the drug information belongs to If the drug information is applicable and does not belong to prohibited drug information, it is determined that the drug information can be taken, that is, the review is passed; if the drug information is applicable drug information and prohibited drug information, the drug information does not belong to applicable drug information and is not prohibited drug information. If the drug information or the drug information does not belong to applicable drug information and belongs to prohibited drug information, it will be determined that the drug information can not be taken, that is, the review will not pass.
  • the drug information does not belong to banned drug information and belongs to applicable drug information, it means that the drug is suitable for the user's condition, is suitable for the user, and will not cause certain harm to the user.
  • the purchase authority for drugs can be granted; if the drug information is not prohibited drug information and does not belong to applicable drug information, it means that the drug is not suitable for the user's condition, and the purchase authority for the drug should not be granted; if the drug information If the information is prohibited drug information and does not belong to applicable drug information, it means that the drug is not suitable for the user, may cause certain harm to the user, and is not suitable for the user's condition, and the purchase authority for the drug should not be granted; if If the drug information is prohibited drug information and applicable drug information, it means that the drug is not suitable for the user and may cause certain harm to the user, and the purchase authority for the drug should not be granted.
  • the reasons for the failed review and the recommended applicable drugs can be displayed to the user; after the review fails because it belongs to the prohibited drug information, The user can be shown the reasons for failure to pass the review, as well as the recommended applicable drugs and the drugs that should be noted as prohibited; after the review fails due to prohibited drug information, the user can be shown the reasons for failure to pass the review, as well as the drugs that should be noted as prohibited. .
  • the user can further extract a request for manual review based on the feedback information, and use an online doctor to review the prescription order.
  • the embodiment of the present application generates prescription orders by generating prescription information from the prescription information recognized by the input information, and by receiving prescription information returned by the user according to the prescription information collection request to generate prescription orders, thereby lowering the threshold for users to purchase drugs online.
  • Improves the convenience of drug purchase obtains banned drugs by analyzing the medical history of the drug user information in the prescription form; then constructs a vector matrix based on the condition information in the prescription form, and uses the condition analysis model to perform model calculations on the vector matrix to obtain applicable drugs information, and then review drug information based on banned drugs and applicable drugs, increasing the diversity of prescription review perspectives and improving the accuracy of prescription review. Therefore, the prescription review method based on artificial intelligence proposed in this application can solve the problem of low accuracy in prescription order review during drug purchase.
  • FIG. 4 it is a functional module diagram of an artificial intelligence-based prescription review device provided by an embodiment of the present application.
  • the artificial intelligence-based prescription review device 100 described in this application can be installed in an electronic device. According to the implemented functions, the artificial intelligence-based prescription review device 100 may include a prescription information judgment module 101, a prescription order generation module 102, a prohibited drug information generation module 103, an applicable drug information generation module 104, and a prescription order review module 105.
  • the module described in this application can also be called a unit, which refers to a series of computer program segments that can be executed by the processor of the electronic device and can complete a fixed function, and are stored in the memory of the electronic device.
  • each module/unit is as follows:
  • the prescription information determination module 101 is used to receive user input information and determine whether there is prescription information in the input information
  • the prescription order generation module 102 is configured to generate a prescription order based on the prescription information when the input information contains prescription information; when there is no prescription information in the input information, send a prescription information collection request to the user; receive The user collects the prescription information returned by the request based on the prescription information collection, and generates a prescription order based on the returned prescription information;
  • the prohibited drug information generation module 103 is used to extract drug user information, condition information and drug information according to the prescription form, and perform medical history analysis based on the drug user information to obtain prohibited drug information;
  • the applicable drug information generation module 104 is used to construct a target vector matrix according to the condition information, and use a pre-built condition analysis model to calculate the target vector matrix to obtain applicable drug information;
  • the prescription review module 105 is configured to review the prescription based on the drug information, the prohibited drug information, and the applicable drug information, and output the review results.
  • each module described in the artificial intelligence-based prescription review device 100 described in the embodiment of the present application adopts the same technical means as the artificial intelligence-based prescription review method described in Figures 1 to 3 above when used. , and can produce the same technical effect, so we will not go into details here.
  • FIG. 5 it is a schematic structural diagram of an electronic device that implements an artificial intelligence-based prescription review method provided by an embodiment of the present application.
  • the electronic device 1 may include a processor 10, a memory 11, a communication bus 12 and a communication interface 13, and may also include a computer program stored in the memory 11 and executable on the processor 10, such as based on artificial intelligence. prescription review process.
  • the processor 10 may be composed of an integrated circuit in some embodiments, for example, it may be composed of a single packaged integrated circuit, or it may be composed of multiple integrated circuits packaged with the same function or different functions, including one or A combination of multiple central processing units (CPUs), microprocessors, digital processing chips, graphics processors and various control chips.
  • the processor 10 is the control core (ControlUnit) of the electronic device. It uses various interfaces and lines to connect various components of the entire electronic device, and runs or executes programs or modules stored in the memory 11 (for example, executing programs based on artificial intelligence prescription review program, etc.), and call the data stored in the memory 11 to perform various functions of the electronic device and process data.
  • ControlUnit ControlUnit
  • the memory 11 includes at least one type of readable storage medium.
  • the readable storage medium includes flash memory, mobile hard disk, multimedia card, card-type memory (such as SD or DX memory, etc.), magnetic memory, magnetic disk, optical disk, etc. .
  • the memory 11 may be an internal storage unit of an electronic device, such as a mobile hard disk of the electronic device.
  • the memory 11 may also be an external storage device of an electronic device, such as a plug-in mobile hard disk, a smart memory card (SmartMediaCard, SMC), or a secure digital (SecureDigital, SD) card equipped on the electronic device. FlashCard, etc.
  • the memory 11 may also include both an internal storage unit of the electronic device and an external storage device.
  • the memory 11 can not only be used to store application software installed on the electronic device and various types of data, such as codes for prescription review programs based on artificial intelligence, etc., but can also be used to temporarily store data that has been output or will be output.
  • the communication bus 12 may be a peripheral component interconnect (PCI) bus or an extended industry standard architecture (EISA) bus.
  • PCI peripheral component interconnect
  • EISA extended industry standard architecture
  • the bus can be divided into address bus, data bus, control bus, etc.
  • the bus is configured to enable connection communication between the memory 11 and at least one processor 10 and the like.
  • the communication interface 13 is used for communication between the above-mentioned electronic device and other devices, and includes a network interface and a user interface.
  • the network interface may include a wired interface and/or a wireless interface (such as a WI-FI interface, a Bluetooth interface, etc.), which are generally used to establish communication connections between the electronic device and other electronic devices.
  • the user interface may be a display (Display) or an input unit (such as a keyboard).
  • the user interface may also be a standard wired interface or a wireless interface.
  • the display may be an LED display, a liquid crystal display, a touch-sensitive liquid crystal display, an OLED (Organic Light-Emitting Diode, organic light-emitting diode) touch device, or the like.
  • the display may also be appropriately referred to as a display screen or a display unit, and is used for displaying information processed in the electronic device and for displaying a visualized user interface.
  • FIG. 5 only shows an electronic device with components. Persons skilled in the art can understand that the structure shown in FIG. 5 does not limit the electronic device 1 and may include fewer or more components than shown in the figure. components, or combinations of certain components, or different arrangements of components.
  • the electronic device may also include a power supply (such as a battery) that supplies power to various components.
  • the power supply may be logically connected to the at least one processor 10 through a power management device, so that the power supply may be logically connected to the at least one processor 10 through a power management device. Realize functions such as charging management, discharge management, and power consumption management.
  • the power supply may also include one or more DC or AC power supplies, recharging devices, power failure detection circuits, power converters or inverters, power status indicators and other arbitrary components.
  • the electronic device may also include a variety of sensors, Bluetooth modules, Wi-Fi modules, etc., which will not be described again here.
  • the artificial intelligence-based prescription review program stored in the memory 11 of the electronic device 1 is a combination of multiple instructions. When run in the processor 10, it can realize:
  • the prescription form is reviewed based on the drug information, the prohibited drug information and the applicable drug information, and the review results are output.
  • the integrated modules/units of the electronic device 1 are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium.
  • the computer-readable storage medium may be volatile or non-volatile.
  • the computer-readable medium may include: any entity or device capable of carrying the computer program code, a recording medium, a USB flash drive, a mobile hard disk, a magnetic disk, an optical disk, a computer memory, and a read-only memory (ROM). ).
  • This application also provides a computer-readable storage medium.
  • the readable storage medium stores a computer program. When executed by a processor of an electronic device, the computer program can realize:
  • the prescription form is reviewed based on the drug information, the prohibited drug information and the applicable drug information, and the review results are output.
  • modules described as separate components may or may not be physically separated, and the components shown as modules may or may not be physical units, that is, they may be located in one place, or they may be distributed to multiple network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the solution of this embodiment.
  • each functional module in various embodiments of the present application can be integrated into one processing unit, or each unit can exist physically alone, or two or more units can be integrated into one unit.
  • the above-mentioned integrated unit can be implemented in the form of hardware or in the form of hardware plus software function modules.
  • Blockchain is a new application model of computer technology such as distributed data storage, point-to-point transmission, consensus mechanism, and encryption algorithm.
  • Blockchain is essentially a decentralized database. It is a series of data blocks generated using cryptographic methods. Each data block contains a batch of network transaction information and is used to verify its Validity of information (anti-counterfeiting) and generation of the next block.
  • Blockchain can include the underlying platform of the blockchain, the platform product service layer, and the application service layer.
  • AI Artificial Intelligence
  • digital computers or machines controlled by digital computers to simulate, extend and expand human intelligence, perceive the environment, acquire knowledge and use knowledge to obtain the best results.

Abstract

一种基于人工智能的处方审核方法、审核装置、设备以及介质,涉及人工智能技术,方法包括:接收用户的输入信息;若输入信息中有处方信息,则根据处方信息生成处方单;若输入信息中没有处方信息,接收用户根据处方信息征集请求返回的处方信息,根据返回的处方信息生成处方单;根据处方单提取用药人信息、病情信息及药品信息,根据用药人信息进行病史分析,得到禁用药品信息;根据病情信息构建目标向量矩阵,利用病情分析模型对目标向量矩阵进行计算,得到适用药品信息;根据药品信息、禁用药品信息和适用药品信息,对处方单进行审核,得到审核结果。该方法可以提高药品购买中处方单的审核精确性。

Description

基于人工智能的处方审核方法、装置、设备及介质
本申请要求于2022年03月23日提交中国专利局、申请号为202210288219.0,发明名称为“基于人工智能的处方审核方法、装置、设备及介质”的中国专利申请的优先权,其全部内容通过引用结合在本申请中。
技术领域
本申请涉及人工智能技术领域,尤其涉及一种基于人工智能的处方审核方法、装置、电子设备及计算机可读存储介质。
背景技术
在国家和地方政策的支持下,互联网医疗欣欣向荣,人民群众逐步接受在线医疗的形式,网上购买药品的消费者越来越多。发明人意识到,在互联网医疗平台售卖药品,例如处方药,也需要像线下医院一样,根据处方单售卖对应的药品,大多数互联网医疗平台的处方药售卖流程是让用户上传处方单照片,利用线上医生或者数据库检索来判断处方单是否合规,进而售卖药品。但是若只将处方单照片来作为药品售卖的受理门槛,会存在没有处方单的用户购买药品受限,进而导致互联网药品售卖的便利性低;若只利用医生在线上进行人工审核,则会导致人工的工作量较大且药品售卖的滞后性高;若利用智能数据库对处方单审核,则可能存在审批角度单一,准确性较低。
技术解决方案
本申请提供的一种基于人工智能的处方审核方法,包括:
接收用户的输入信息,判断所述输入信息中是否有处方信息;
若所述输入信息中有处方信息,则根据所述处方信息生成处方单;
若所述输入信息中没有处方信息,则向用户发送处方信息征集请求;
接收用户根据所述处方信息征集请求返回的处方信息,并根据所述返回的处方信息生成处方单;
根据所述处方单提取用药人信息、病情信息及药品信息,并根据所述用药人信息进行病史分析,得到禁用药品信息;
根据所述病情信息构建目标向量矩阵,并利用预构建的病情分析模型对所述目标向量矩阵进行计算,得到适用药品信息;
根据所述药品信息、所述禁用药品信息以及所述适用药品信息,对所述处方单进行审核,并输出审核结果。
本申请还提供一种基于人工智能的处方审核装置,所述装置包括:
处方信息判断模块,用于接收用户的输入信息,判断所述输入信息中是否有处方信息;
处方单生成模块,用于在所述输入信息中有处方信息时,根据所述处方信息生成处方单;在所述输入信息中没有处方信息时,向用户发送处方信息征集请求;接收用户根据所述处方信息征集请求返回的处方信息,并根据所述返回的处方信息生成处方单;
禁用药品信息生成模块,用于根据所述处方单提取用药人信息、病情信息及药品信息,并根据所述用药人信息进行病史分析,得到禁用药品信息;
适用药品信息生成模块,用于根据所述病情信息构建目标向量矩阵,并利用预构建的病情分析模型对所述目标向量矩阵进行计算,得到适用药品信息;
处方单审核模块,用于根据所述药品信息、所述禁用药品信息以及所述适用药品信息,对所述处方单进行审核,并输出审核结果。
本申请还提供一种电子设备,所述电子设备包括:
至少一个处理器;以及,
与所述至少一个处理器通信连接的存储器;其中,
所述存储器存储有可被所述至少一个处理器执行的计算机程序,所述计算机程序被所述至少一个处理器执行,以使所述至少一个处理器能够执行如下所述的基于人工智能的处方审核方法:
接收用户的输入信息,判断所述输入信息中是否有处方信息;
若所述输入信息中有处方信息,则根据所述处方信息生成处方单;
若所述输入信息中没有处方信息,则向用户发送处方信息征集请求;
接收用户根据所述处方信息征集请求返回的处方信息,并根据所述返回的处方信息生成处方单;
根据所述处方单提取用药人信息、病情信息及药品信息,并根据所述用药人信息进行病史分析,得到禁用药品信息;
根据所述病情信息构建目标向量矩阵,并利用预构建的病情分析模型对所述目标向量矩阵进行计算,得到适用药品信息;
根据所述药品信息、所述禁用药品信息以及所述适用药品信息,对所述处方单进行审核,并输出审核结果。
本申请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中存储有至少一个计算机程序,所述至少一个计算机程序被电子设备中的处理器执行以实现如下所述的基于人工智能的处方审核方法:
接收用户的输入信息,判断所述输入信息中是否有处方信息;
若所述输入信息中有处方信息,则根据所述处方信息生成处方单;
若所述输入信息中没有处方信息,则向用户发送处方信息征集请求;
接收用户根据所述处方信息征集请求返回的处方信息,并根据所述返回的处方信息生成处方单;
根据所述处方单提取用药人信息、病情信息及药品信息,并根据所述用药人信息进行病史分析,得到禁用药品信息;
根据所述病情信息构建目标向量矩阵,并利用预构建的病情分析模型对所述目标向量矩阵进行计算,得到适用药品信息;
根据所述药品信息、所述禁用药品信息以及所述适用药品信息,对所述处方单进行审核,并输出审核结果。
附图说明
图1为本申请一实施例提供的基于人工智能的处方审核方法的流程示意图;
图2为本申请一实施例提供的返回的处方信息生成处方单的流程示意图;
图3为本申请一实施例提供的进行病史分析的流程示意图;
图4为本申请一实施例提供的基于人工智能的处方审核装置的功能模块图;
图5为本申请一实施例提供的实现所述基于人工智能的处方审核方法的电子设备的结构示意图。
本申请目的的实现、功能特点及优点将结合实施例,参照附图做进一步说明。
本发明的实施方式
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。
本申请实施例提供一种基于人工智能的处方审核方法。所述基于人工智能的处方审核方法的执行主体包括但不限于服务端、终端等能够被配置为执行本申请实施例提供的该方法的电子设备中的至少一种。换言之,所述基于人工智能的处方审核方法可以由安装在终端设备或服务端设备的软件或硬件来执行,所述软件可以是区块链平台。所述服务端包括但不限于:单台服务器、服务器集群、云端服务器或云端服务器集群等。所述服务器可以是独立的服务器,也可以是提供云服务、云数据库、云计算、云函数、云存储、网络服务、云通信、中间件服务、域名服务、安全服务、内容分发网络(ContentDeliveryNetwork,CDN)、以及大数据和人工智能平台等基础云计算服务的云服务器。
参照图1所示,为本申请一实施例提供的基于人工智能的处方审核方法的流程示意图。在本实施例中,所述基于人工智能的处方审核方法包括:
S1、接收用户的输入信息,判断所述输入信息中是否有处方信息;
本申请实施例中,所述用户的输入信息可以为医院、诊所开具的处方单图片、其他渠道导入的含有处方信息的表格等。
本申请实施例中,所述判断所述输入信息中是否有处方信息,包括:
提取所述输入信息的信息格式,并判断所述信息格式是否包含预设格式;
若所述信息格式包含预设格式,则判定所述输入信息中有处方信息;
若所述信息格式不包含预设格式,则判定所述输入信息中没有处方信息。
详细地,所述信息格式可以为JPG、BMP、PNG、EXCEL等;若预设信息格式为JPG、BMP、PNG、EXCEL,当输入信息中存在信息格式为JPG、BMP、PNG等时,则所述信息格式包含预设格式,即输入信息中含有处方单图像,该处方单图像即为处方信息。
若所述输入信息中有处方信息,则执行S2、根据所述处方信息生成处方单;
本申请实施例中,所述处方信息可以为处方单图像等包含处方信息的显示形式。
本申请实施例中,所述根据所述处方信息生成处方单,包括:
对所述处方信息进行文本识别,得到处方文本;
根据所述处方文本,利用预设的处方单模板生成处方单。
进一步地,所述对所述处方信息进行文本识别,得到处方文本,包括:
对所述处方信息进行文本类型识别,得到所述处方信息对应的文本类型;
根据所述文本类型从预设的文本定位模型库中选择对应的文本定位模型作为目标文本定位模型;
利用所述目标文本定位模型对所述处方信息进行文本定位,得到文本区域;
对所述文本区域进行光学字符文本识别,得到处方文本。
本申请实施例中,所述处方信息可以为处方单图像,文本类型包含有印刷体文本类型和手写体文本类型;可以利用图像分类模型对所述处方信息进行文本类型识别,所述图像分类模型可以是基于ResNet18算法的深度学习模型;所述目标文本图像对应的图像文本类型可以是目标文本图像的待识别文本区域对应的图像文本类型,也可以是整个目标文本图像对应的图像文本类型
本申请实施例中,可以利用光学字符文本识别模型对所述文本区域进行光学字符文本识别。若所述文本类型为手写体文本类型,则将所述待识别文本区域输入预设的手写体光学字符文本识别模型;若所述文本类型为印刷体文本类型,则将所述待识别文本区域输入预设的印刷体光学字符文本识别模型。所述手写体光学字符文本识别模型和印刷体光学字符文本识别模型均可以为卷积循环神经网络(CRNN)模型。
若所述输入信息中没有处方信息,则执行S3、向用户发送处方信息征集请求;
本申请实施例中,若所述输入信息中没有处方信息,则需要用户进行主动输入,以生成处方单,因此需要向用户发送处方信息征集请求以提醒用户进行主动输入。
S4、接收用户根据所述处方信息征集请求返回的处方信息,并根据所述返回的处方信息生成处方单;
本申请实施例中,所述用户根据所述处方信息征集请求返回的处方信息可以为补充的医院、诊所开具的处方单图片、其他渠道导入的含有处方信息的表格,或者用户根据处方信息征集请求附带的模板所填写的个人病历情况。
本申请实施例中,请参阅图2所示,所述根据所述返回的处方信息生成处方单,包括:
S41、对所述返回的处方信息进行语义分析,得到多个语义段落;
S42、根据所述多个语义段落与预设的处方单模板的模板段落进行相似度计算;
S43、将相似度计算结果大于阈值的语义段落输入对应的模板段落中,得到处方单。
本申请实施例中,可以利用自然语言学习(NPL)中的深度学习文本分类模型(例如fastText模型、TextCNN、TextRNN 、TextRNN + Attention、TextRCNN等)对所述返回的处方信息进行语义分析。
本申请实施例中,可以利用余弦相似度、皮尔逊相关系数、欧氏距离等计算多个语义段落与预设的处方单模板的模板段落的相似度。
本申请其中一个实际应用场景中,向用户发送处方信息征集请求时可以附带有处方单填写模板,用户可以直接根据该处方单模板进行信息输入,输入的信息即为处方信息,用户返回该处方信息后,可以结果处方单模板直接生成处方单。
S5、根据所述处方单提取用药人信息、病情信息及药品信息,并根据所述用药人信息进行病史分析,得到禁用药品信息;
本申请实施例中,所述用药人信息可以包括实名认证的个人信息、过往病史、药物过敏史、家族遗传病等信息,所述病情信息可以为用户针对病情的描述,所述药品信息可以为用户想要购买的具体药品名称等。
本申请实施例中,所述处方单可以为具有固定格式模板的文档、表格等,通过提取固定位置的内容即可以获取相应的用药人信息、病情信息或者药品信息。
本申请实施例中,所述禁用药品信息为通过对用药人的过往病史、个人信息等得到禁止或注意服用、使用的药品。
本申请实施例中,请参阅图3所示,所述根据所述用药人信息进行病史分析,得到禁用药品信息,包括:
S51、对所述用药人信息进行分词处理,得到多个信息分词;
S52、利用预设的禁用药信息库对所述多个信息分词逐一检索;
S53、根据选取检索到的信息分词从所述禁用药信息库中提取对应的禁用药,得到禁用药信息。
本申请实施例中,所述禁用药信息库中包含着不同病史以及对应的禁用药,不同年龄段对应的禁用药等。
例如,若存在信息分词 “糖尿病”,该分词在禁用药信息库中可以检索到,并且对应的禁用药包括泼尼松、地塞米松、倍他米松等,则所述禁用药品信息包括泼尼松、地塞米松、倍他米松;若存在信息分词“5岁”,该分词在禁用药信息库中可以检索到,并且对应的禁用药包括羟氯喹、丙咪嗪、雷尼替丁等,则所述禁用药品信息包括羟氯喹、丙咪嗪、雷尼替丁。
S6、根据所述病情信息构建目标向量矩阵,并利用预构建的病情分析模型对所述目标向量矩阵进行计算,得到适用药品信息;
本申请实施例中,所述适用药品信息为根据病情信息得到的应该使用的药品信息。
本申请实施例中,所述根据所述病情信息构建目标向量矩阵,包括:
从所述病情信息中提取待掩码数据,对所述待掩码数据执行掩码操作,得到已掩码数据;
对所述已掩码数据中的所有数据进行向量转换,得到向量集,并对所述向量集执行位置编码,得到定位向量集;
将所述定位向量集转换为定位向量矩阵, 利用所述定位向量矩阵调节预构建的前馈神经网络中的迭代权重因子,得到目标向量矩阵。
本申请实施例中,可以根据预设的掩码概率,从所述待掩码数据中提取关键字,对所述关键字执行掩码操作,得到已掩码字;在所述待掩码数据中,用所述已掩码字替换所述关键字,得到所述已掩码数据。
本申请实施例中,可采用Word2vec算法,将所述已掩码数据中的所有数据进行向量转换。
本申请实施例中,所述利用预构建的病情分析模型对所述目标向量矩阵进行计算,得到适用药品信息之前,所述方法还包括:
获取训练病情信息以及所述训练病情信息对应的真实适应药品信息;
构建所述训练病情信息的向量矩阵,将所述向量矩阵输入预构建的病情分析模型,得到所述病情分析模型的输出结果;
利用预设的疾病名称损失函数计算得到所述输出结果与所述真实适应药品信息的损失值,根据所述损失值优化所述病情分析模型,得到标准病情分析模型。
本申请实施例中,所述病情分析模型是一种预训练语言模型,包括但不限于BERT模型(BidirectionalEncoderRepresentationsfromTransformers,双向编码表示模型)、LSTM模型(Long-ShortTermMemory, 长短期记忆模型)。
本申请实施例中,所述利用预构建的病情分析模型对所述目标向量矩阵进行计算,得到适用药品信息,包括:
通过所述病情分析模型对所述目标向量矩阵进行预设次数的卷积、池化和全连接,得到病情分析信息;
通过激活器计算得到所述病情分析信息对应的适用药品信息。
例如,假设病情描述为“头疼、低烧”,将其输入病情分析模型,输出结果为乙酰氨基酚、布洛芬、阿司匹林,则该输出结果即可作为适用药品信息。
S7、根据所述药品信息、所述禁用药品信息以及所述适用药品信息,对所述处方单进行审核,并输出审核结果。
本申请实施例中,所述根据所述药品信息、所述禁用药品信息以及所述适用药品信息,对所述处方单进行审核,并输出审核结果,包括:
判断所述药品信息是否不属于禁用药品信息且属于适用药品信息;
若所述药品信息不属于禁用药品信息且属于适用药品信息,则执行判定所述处方单审核通过;
若所述药品信息不属于禁用药品信息且不属于适用药品信息,则执行判定所述处方单审核不通过;
若所述药品信息属于禁用药品信息且不属于适用药品信息,则执行判定所述处方单审核不通过;
若所述药品信息属于禁用药品信息且属于适用药品信息,则执行判定所述处方单审核不通过。
本申请实施例中,通过分别对用药人信息和病情信息的分析,得到禁用药信息和适用药信息,已划分出药品范围,只要根据所述药品信息进行信息比对,在所述药品信息属于适用药信息且不属于禁用药信息,则判定该药品信息可以服用,即审核通过;在所述药品信息属于适用药信息且属于禁用药信息、所述药品信息不属于适用药信息且不属于禁用药信息或者所述药品信息不属于适用药信息且属于禁用药信息,则判定该药品信息可以不服用,即审核不通过。
本申请实施例中,若所述药品信息不属于禁用药品信息且属于适用药品信息,则说明所述药品并适用于用户的病情情况,且适用于该用户,不会对用户造成一定的危害,可以授予药品的购买权限;若所述药品信息不属于禁用药品信息且不属于适用药品信息,则说明所述药品并不适用于用户的病情情况,不应授予药品的购买权限;若所述药品信息属于禁用药品信息且不属于适用药品信息,则说明所述药品不适用于该用户,可能会对用户造成一定的危害,并且不适用于用户的病情情况,不应授予药品的购买权限;若所述药品信息属于禁用药品信息且属于适用药品信息,则说明所述药品不适用于该用户,可能会对用户造成一定的危害,不应授予药品的购买权限。
本申请一可选实施例中,在由于不属于适用药信息而审核不通过后,可以向用户显示审核不通过的原因,以及建议的适用药;在由于属于禁用药信息而审核不通过后,可以向用户显示审核不通过的原因,以及建议的适用药和注意禁止适用的药品;在由于属于禁用药信息而审核不通过后,可以向用户显示审核不通过的原因,以及注意禁止适用的药品。
本申请另一可选实施例中,在审核不通过后,用户可以根据反馈信息进一步提取人工审核的请求,利用线上医生进行处方单的审核。
本申请实施例通过对输入信息识别到的处方信息生成处方单,以及通过接收用户根据处方信息征集请求返回的处方信息生成处方单这两种方式生成处方单,降低了用户在线购买药品的门槛,提高了药品购买的便利性;通过对处方单中的用药人信息进行病史分析得到禁用药品;再根据处方单中病情信息构建向量矩阵,利用病情分析模型对该向量矩阵进行模型计算,得到适用药品信息,进而根据禁用药和适用药对药品信息进行审核,增加了处方单审核角度的多样性,提高了处方单审核的精确性。因此本申请提出的基于人工智能的处方审核方法,可以解决药品购买中处方单审核精确性较低的问题。
如图4所示,是本申请一实施例提供的基于人工智能的处方审核装置的功能模块图。
本申请所述基于人工智能的处方审核装置100可以安装于电子设备中。根据实现的功能,所述基于人工智能的处方审核装置100可以包括处方信息判断模块101、处方单生成模块102、禁用药品信息生成模块103、适用药品信息生成模块104及处方单审核模块105。本申请所述模块也可以称之为单元,是指一种能够被电子设备处理器所执行,并且能够完成固定功能的一系列计算机程序段,其存储在电子设备的存储器中。
在本实施例中,关于各模块/单元的功能如下:
所述处方信息判断模块101,用于接收用户的输入信息,判断所述输入信息中是否有处方信息;
所述处方单生成模块102,用于在所述输入信息中有处方信息时,根据所述处方信息生成处方单;在所述输入信息中没有处方信息时,向用户发送处方信息征集请求;接收用户根据所述处方信息征集请求返回的处方信息,并根据所述返回的处方信息生成处方单;
所述禁用药品信息生成模块103,用于根据所述处方单提取用药人信息、病情信息及药品信息,并根据所述用药人信息进行病史分析,得到禁用药品信息;
所述适用药品信息生成模块104,用于根据所述病情信息构建目标向量矩阵,并利用预构建的病情分析模型对所述目标向量矩阵进行计算,得到适用药品信息;
所述处方单审核模块105,用于根据所述药品信息、所述禁用药品信息以及所述适用药品信息,对所述处方单进行审核,并输出审核结果。
详细地,本申请实施例中所述基于人工智能的处方审核装置100中所述的各模块在使用时采用与上述图1至图3中所述的基于人工智能的处方审核方法一样的技术手段,并能够产生相同的技术效果,这里不再赘述。
如图5所示,是本申请一实施例提供的实现基于人工智能的处方审核方法的电子设备的结构示意图。
所述电子设备1可以包括处理器10、存储器11、通信总线12以及通信接口13,还可以包括存储在所述存储器11中并可在所述处理器10上运行的计算机程序,如基于人工智能的处方审核程序。
其中,所述处理器10在一些实施例中可以由集成电路组成,例如可以由单个封装的集成电路所组成,也可以是由多个相同功能或不同功能封装的集成电路所组成,包括一个或者多个中央处理器(CentralProcessingunit,CPU)、微处理器、数字处理芯片、图形处理器及各种控制芯片的组合等。所述处理器10是所述电子设备的控制核心(ControlUnit),利用各种接口和线路连接整个电子设备的各个部件,通过运行或执行存储在所述存储器11内的程序或者模块(例如执行基于人工智能的处方审核程序等),以及调用存储在所述存储器11内的数据,以执行电子设备的各种功能和处理数据。
所述存储器11至少包括一种类型的可读存储介质,所述可读存储介质包括闪存、移动硬盘、多媒体卡、卡型存储器(例如:SD或DX存储器等)、磁性存储器、磁盘、光盘等。所述存储器11在一些实施例中可以是电子设备的内部存储单元,例如该电子设备的移动硬盘。所述存储器11在另一些实施例中也可以是电子设备的外部存储设备,例如电子设备上配备的插接式移动硬盘、智能存储卡(SmartMediaCard, SMC)、安全数字(SecureDigital, SD)卡、闪存卡(FlashCard)等。进一步地,所述存储器11还可以既包括电子设备的内部存储单元也包括外部存储设备。所述存储器11不仅可以用于存储安装于电子设备的应用软件及各类数据,例如基于人工智能的处方审核程序的代码等,还可以用于暂时地存储已经输出或者将要输出的数据。
所述通信总线12可以是外设部件互连标准(peripheralcomponentinterconnect,简称PCI)总线或扩展工业标准结构(extendedindustrystandardarchitecture,简称EISA)总线等。该总线可以分为地址总线、数据总线、控制总线等。所述总线被设置为实现所述存储器11以及至少一个处理器10等之间的连接通信。
所述通信接口13用于上述电子设备与其他设备之间的通信,包括网络接口和用户接口。可选地,所述网络接口可以包括有线接口和/或无线接口(如WI-FI接口、蓝牙接口等),通常用于在该电子设备与其他电子设备之间建立通信连接。所述用户接口可以是显示器(Display)、输入单元(比如键盘(Keyboard)),可选地,用户接口还可以是标准的有线接口、无线接口。可选地,在一些实施例中,显示器可以是LED显示器、液晶显示器、触控式液晶显示器以及OLED(OrganicLight-EmittingDiode,有机发光二极管)触摸器等。其中,显示器也可以适当的称为显示屏或显示单元,用于显示在电子设备中处理的信息以及用于显示可视化的用户界面。
图5仅示出了具有部件的电子设备,本领域技术人员可以理解的是,图5示出的结构并不构成对所述电子设备1的限定,可以包括比图示更少或者更多的部件,或者组合某些部件,或者不同的部件布置。
例如,尽管未示出,所述电子设备还可以包括给各个部件供电的电源(比如电池),优选地,电源可以通过电源管理装置与所述至少一个处理器10逻辑相连,从而通过电源管理装置实现充电管理、放电管理、以及功耗管理等功能。电源还可以包括一个或一个以上的直流或交流电源、再充电装置、电源故障检测电路、电源转换器或者逆变器、电源状态指示器等任意组件。所述电子设备还可以包括多种传感器、蓝牙模块、Wi-Fi模块等,在此不再赘述。
应该了解,所述实施例仅为说明之用,在专利申请范围上并不受此结构的限制。
所述电子设备1中的所述存储器11存储的基于人工智能的处方审核程序是多个指令的组合,在所述处理器10中运行时,可以实现:
接收用户的输入信息,判断所述输入信息中是否有处方信息;
若所述输入信息中有处方信息,则根据所述处方信息生成处方单;
若所述输入信息中没有处方信息,则向用户发送处方信息征集请求;
接收用户根据所述处方信息征集请求返回的处方信息,并根据所述返回的处方信息生成处方单;
根据所述处方单提取用药人信息、病情信息及药品信息,并根据所述用药人信息进行病史分析,得到禁用药品信息;
根据所述病情信息构建目标向量矩阵,并利用预构建的病情分析模型对所述目标向量矩阵进行计算,得到适用药品信息;
根据所述药品信息、所述禁用药品信息以及所述适用药品信息,对所述处方单进行审核,并输出审核结果。
具体地,所述处理器10对上述指令的具体实现方法可参考附图对应实施例中相关步骤的描述,在此不赘述。
进一步地,所述电子设备1集成的模块/单元如果以软件功能单元的形式实现并作为独立的产品销售或使用时,可以存储在一个计算机可读存储介质中。所述计算机可读存储介质可以是易失性的,也可以是非易失性的。例如,所述计算机可读介质可以包括:能够携带所述计算机程序代码的任何实体或装置、记录介质、U盘、移动硬盘、磁碟、光盘、计算机存储器、只读存储器(ROM,Read-OnlyMemory)。
本申请还提供一种计算机可读存储介质,所述可读存储介质存储有计算机程序,所述计算机程序在被电子设备的处理器所执行时,可以实现:
接收用户的输入信息,判断所述输入信息中是否有处方信息;
若所述输入信息中有处方信息,则根据所述处方信息生成处方单;
若所述输入信息中没有处方信息,则向用户发送处方信息征集请求;
接收用户根据所述处方信息征集请求返回的处方信息,并根据所述返回的处方信息生成处方单;
根据所述处方单提取用药人信息、病情信息及药品信息,并根据所述用药人信息进行病史分析,得到禁用药品信息;
根据所述病情信息构建目标向量矩阵,并利用预构建的病情分析模型对所述目标向量矩阵进行计算,得到适用药品信息;
根据所述药品信息、所述禁用药品信息以及所述适用药品信息,对所述处方单进行审核,并输出审核结果。
在本申请所提供的几个实施例中,应该理解到,所揭露的设备,装置和方法,可以通过其它的方式实现。例如,以上所描述的装置实施例仅仅是示意性的,例如,所述模块的划分,仅仅为一种逻辑功能划分,实际实现时可以有另外的划分方式。
所述作为分离部件说明的模块可以是或者也可以不是物理上分开的,作为模块显示的部件可以是或者也可以不是物理单元,即可以位于一个地方,或者也可以分布到多个网络单元上。可以根据实际的需要选择其中的部分或者全部模块来实现本实施例方案的目的。
另外,在本申请各个实施例中的各功能模块可以集成在一个处理单元中,也可以是各个单元单独物理存在,也可以两个或两个以上单元集成在一个单元中。上述集成的单元既可以采用硬件的形式实现,也可以采用硬件加软件功能模块的形式实现。
对于本领域技术人员而言,显然本申请不限于上述示范性实施例的细节,而且在不背离本申请的精神或基本特征的情况下,能够以其他的具体形式实现本申请。
因此,无论从哪一点来看,均应将实施例看作是示范性的,而且是非限制性的,本申请的范围由所附权利要求而不是上述说明限定,因此旨在将落在权利要求的等同要件的含义和范围内的所有变化涵括在本申请内。不应将权利要求中的任何附关联图标记视为限制所涉及的权利要求。
本申请所指区块链是分布式数据存储、点对点传输、共识机制、加密算法等计算机技术的新型应用模式。区块链(Blockchain),本质上是一个去中心化的数据库,是一串使用密码学方法相关联产生的数据块,每一个数据块中包含了一批次网络交易的信息,用于验证其信息的有效性(防伪)和生成下一个区块。区块链可以包括区块链底层平台、平台产品服务层以及应用服务层等。
本申请实施例可以基于人工智能技术对相关的数据进行获取和处理。其中,人工智能(ArtificialIntelligence,AI)是利用数字计算机或者数字计算机控制的机器模拟、延伸和扩展人的智能,感知环境、获取知识并使用知识获得最佳结果的理论、方法、技术及应用系统。
此外,显然“包括”一词不排除其他单元或步骤,单数不排除复数。系统权利要求中陈述的多个单元或装置也可以由一个单元或装置通过软件或者硬件来实现。第一、第二等词语用来表示名称,而并不表示任何特定的顺序。
最后应说明的是,以上实施例仅用以说明本申请的技术方案而非限制,尽管参照较佳实施例对本申请进行了详细说明,本领域的普通技术人员应当理解,可以对本申请的技术方案进行修改或等同替换,而不脱离本申请技术方案的精神和范围。

Claims (20)

  1. 一种基于人工智能的处方审核方法,其中,所述方法包括:
    接收用户的输入信息,判断所述输入信息中是否有处方信息;
    若所述输入信息中有处方信息,则根据所述处方信息生成处方单;
    若所述输入信息中没有处方信息,则向用户发送处方信息征集请求;
    接收用户根据所述处方信息征集请求返回的处方信息,并根据所述返回的处方信息生成处方单;
    根据所述处方单提取用药人信息、病情信息及药品信息,并根据所述用药人信息进行病史分析,得到禁用药品信息;
    根据所述病情信息构建目标向量矩阵,并利用预构建的病情分析模型对所述目标向量矩阵进行计算,得到适用药品信息;
    根据所述药品信息、所述禁用药品信息以及所述适用药品信息,对所述处方单进行审核,并输出审核结果。
  2. 如权利要求1所述的基于人工智能的处方审核方法,其中,所述判断所述输入信息中是否有处方信息,包括:
    提取所述输入信息的信息格式,并判断所述信息格式是否包含预设格式;
    若所述信息格式包含预设格式,则判定所述输入信息中有处方信息;
    若所述信息格式不包含预设格式,则判定所述输入信息中没有处方信息。
  3. 如权利要求1所述的基于人工智能的处方审核方法,其中,所述根据所述返回的处方信息生成处方单,包括:
    对所述返回的处方信息进行语义分析,得到多个语义段落;
    根据所述多个语义段落与预设的处方单模板的模板段落进行相似度计算;
    将相似度计算结果大于阈值的语义段落输入对应的模板段落中,得到处方单。
  4. 如权利要求1所述的基于人工智能的处方审核方法,其中,所述根据所述用药人信息进行病史分析,得到禁用药品信息,包括:
    对所述用药人信息进行分词处理,得到多个信息分词;
    利用预设的禁用药信息库对所述多个信息分词逐一检索;
    根据选取检索到的信息分词从所述禁用药信息库中提取对应的禁用药,得到禁用药信息。
  5. 如权利要求1所述的基于人工智能的处方审核方法,其中,所述根据所述病情信息构建目标向量矩阵,包括:
    从所述病情信息中提取待掩码数据,对所述待掩码数据执行掩码操作,得到已掩码数据;
    对所述已掩码数据中的所有数据进行向量转换,得到向量集,并对所述向量集执行位置编码,得到定位向量集;
    将所述定位向量集转换为定位向量矩阵,利用所述定位向量矩阵调节预构建的前馈神经网络中的迭代权重因子,得到目标向量矩阵。
  6. 如权利要求1所述的基于人工智能的处方审核方法,其中,所述利用预构建的病情分析模型对所述目标向量矩阵进行计算,得到适用药品信息,包括:
    通过所述病情分析模型对所述目标向量矩阵进行预设次数的卷积、池化和全连接,得到病情分析信息;
    通过激活器计算得到所述病情分析信息对应的适用药品信息。
  7. 如权利要求1至6中任一项所述的基于人工智能的处方审核方法,其中,所述根据所述药品信息、所述禁用药品信息以及所述适用药品信息,对所述处方单进行审核,并输出审核结果,包括:
    判断所述药品信息是否不属于禁用药品信息且属于适用药品信息;
    若所述药品信息不属于禁用药品信息且属于适用药品信息,则执行判定所述处方单审核通过;
    若所述药品信息不属于禁用药品信息且不属于适用药品信息,则执行判定所述处方单审核不通过;
    若所述药品信息属于禁用药品信息且不属于适用药品信息,则执行判定所述处方单审核不通过;
    若所述药品信息属于禁用药品信息且属于适用药品信息,则执行判定所述处方单审核不通过。
  8. 一种基于人工智能的处方审核装置,其中,所述装置包括:
    处方信息判断模块,用于接收用户的输入信息,判断所述输入信息中是否有处方信息;
    处方单生成模块,用于在所述输入信息中有处方信息时,根据所述处方信息生成处方单;在所述输入信息中没有处方信息时,向用户发送处方信息征集请求;接收用户根据所述处方信息征集请求返回的处方信息,并根据所述返回的处方信息生成处方单;
    禁用药品信息生成模块,用于根据所述处方单提取用药人信息、病情信息及药品信息,并根据所述用药人信息进行病史分析,得到禁用药品信息;
    适用药品信息生成模块,用于根据所述病情信息构建目标向量矩阵,并利用预构建的病情分析模型对所述目标向量矩阵进行计算,得到适用药品信息;
    处方单审核模块,用于根据所述药品信息、所述禁用药品信息以及所述适用药品信息,对所述处方单进行审核,并输出审核结果。
  9. 一种电子设备,其中,所述电子设备包括:
    至少一个处理器;以及,
    与所述至少一个处理器通信连接的存储器;其中,
    所述存储器存储有可被所述至少一个处理器执行的计算机程序,所述计算机程序被所述至少一个处理器执行,以使所述至少一个处理器能够执行如下所述的基于人工智能的处方审核方法:
    接收用户的输入信息,判断所述输入信息中是否有处方信息;
    若所述输入信息中有处方信息,则根据所述处方信息生成处方单;
    若所述输入信息中没有处方信息,则向用户发送处方信息征集请求;
    接收用户根据所述处方信息征集请求返回的处方信息,并根据所述返回的处方信息生成处方单;
    根据所述处方单提取用药人信息、病情信息及药品信息,并根据所述用药人信息进行病史分析,得到禁用药品信息;
    根据所述病情信息构建目标向量矩阵,并利用预构建的病情分析模型对所述目标向量矩阵进行计算,得到适用药品信息;
    根据所述药品信息、所述禁用药品信息以及所述适用药品信息,对所述处方单进行审核,并输出审核结果。
  10. 如权利要求9所述的电子设备,其中,所述判断所述输入信息中是否有处方信息,包括:
    提取所述输入信息的信息格式,并判断所述信息格式是否包含预设格式;
    若所述信息格式包含预设格式,则判定所述输入信息中有处方信息;
    若所述信息格式不包含预设格式,则判定所述输入信息中没有处方信息。
  11. 如权利要求9所述的电子设备,其中,所述根据所述返回的处方信息生成处方单,包括:
    对所述返回的处方信息进行语义分析,得到多个语义段落;
    根据所述多个语义段落与预设的处方单模板的模板段落进行相似度计算;
    将相似度计算结果大于阈值的语义段落输入对应的模板段落中,得到处方单。
  12. 如权利要求9所述的电子设备,其中,所述根据所述用药人信息进行病史分析,得到禁用药品信息,包括:
    对所述用药人信息进行分词处理,得到多个信息分词;
    利用预设的禁用药信息库对所述多个信息分词逐一检索;
    根据选取检索到的信息分词从所述禁用药信息库中提取对应的禁用药,得到禁用药信息。
  13. 如权利要求9所述的电子设备,其中,所述根据所述病情信息构建目标向量矩阵,包括:
    从所述病情信息中提取待掩码数据,对所述待掩码数据执行掩码操作,得到已掩码数据;
    对所述已掩码数据中的所有数据进行向量转换,得到向量集,并对所述向量集执行位置编码,得到定位向量集;
    将所述定位向量集转换为定位向量矩阵,利用所述定位向量矩阵调节预构建的前馈神经网络中的迭代权重因子,得到目标向量矩阵。
  14. 如权利要求9所述的电子设备,其中,所述利用预构建的病情分析模型对所述目标向量矩阵进行计算,得到适用药品信息,包括:
    通过所述病情分析模型对所述目标向量矩阵进行预设次数的卷积、池化和全连接,得到病情分析信息;
    通过激活器计算得到所述病情分析信息对应的适用药品信息。
  15. 一种计算机可读存储介质,存储有计算机程序,其中,所述计算机程序被处理器执行时实现如下所述的基于人工智能的处方审核方法:
    接收用户的输入信息,判断所述输入信息中是否有处方信息;
    若所述输入信息中有处方信息,则根据所述处方信息生成处方单;
    若所述输入信息中没有处方信息,则向用户发送处方信息征集请求;
    接收用户根据所述处方信息征集请求返回的处方信息,并根据所述返回的处方信息生成处方单;
    根据所述处方单提取用药人信息、病情信息及药品信息,并根据所述用药人信息进行病史分析,得到禁用药品信息;
    根据所述病情信息构建目标向量矩阵,并利用预构建的病情分析模型对所述目标向量矩阵进行计算,得到适用药品信息;
    根据所述药品信息、所述禁用药品信息以及所述适用药品信息,对所述处方单进行审核,并输出审核结果。
  16. 如权利要求15所述的计算机可读存储介质,其中,所述判断所述输入信息中是否有处方信息,包括:
    提取所述输入信息的信息格式,并判断所述信息格式是否包含预设格式;
    若所述信息格式包含预设格式,则判定所述输入信息中有处方信息;
    若所述信息格式不包含预设格式,则判定所述输入信息中没有处方信息。
  17. 如权利要求15所述的计算机可读存储介质,其中,所述根据所述返回的处方信息生成处方单,包括:
    对所述返回的处方信息进行语义分析,得到多个语义段落;
    根据所述多个语义段落与预设的处方单模板的模板段落进行相似度计算;
    将相似度计算结果大于阈值的语义段落输入对应的模板段落中,得到处方单。
  18. 如权利要求15所述的计算机可读存储介质,其中,所述根据所述用药人信息进行病史分析,得到禁用药品信息,包括:
    对所述用药人信息进行分词处理,得到多个信息分词;
    利用预设的禁用药信息库对所述多个信息分词逐一检索;
    根据选取检索到的信息分词从所述禁用药信息库中提取对应的禁用药,得到禁用药信息。
  19. 如权利要求15所述的计算机可读存储介质,其中,所述根据所述病情信息构建目标向量矩阵,包括:
    从所述病情信息中提取待掩码数据,对所述待掩码数据执行掩码操作,得到已掩码数据;
    对所述已掩码数据中的所有数据进行向量转换,得到向量集,并对所述向量集执行位置编码,得到定位向量集;
    将所述定位向量集转换为定位向量矩阵,利用所述定位向量矩阵调节预构建的前馈神经网络中的迭代权重因子,得到目标向量矩阵。
  20. 如权利要求15所述的计算机可读存储介质,其中,所述利用预构建的病情分析模型对所述目标向量矩阵进行计算,得到适用药品信息,包括:
    通过所述病情分析模型对所述目标向量矩阵进行预设次数的卷积、池化和全连接,得到病情分析信息;
    通过激活器计算得到所述病情分析信息对应的适用药品信息。
PCT/CN2022/122999 2022-03-23 2022-09-30 基于人工智能的处方审核方法、装置、设备及介质 WO2023178978A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202210288219.0A CN114550870A (zh) 2022-03-23 2022-03-23 基于人工智能的处方审核方法、装置、设备及介质
CN202210288219.0 2022-03-23

Publications (1)

Publication Number Publication Date
WO2023178978A1 true WO2023178978A1 (zh) 2023-09-28

Family

ID=81666465

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2022/122999 WO2023178978A1 (zh) 2022-03-23 2022-09-30 基于人工智能的处方审核方法、装置、设备及介质

Country Status (2)

Country Link
CN (1) CN114550870A (zh)
WO (1) WO2023178978A1 (zh)

Families Citing this family (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114550870A (zh) * 2022-03-23 2022-05-27 康键信息技术(深圳)有限公司 基于人工智能的处方审核方法、装置、设备及介质
CN115719628A (zh) * 2022-11-16 2023-02-28 联仁健康医疗大数据科技股份有限公司 一种中医处方生成方法、装置、设备及存储介质
CN116631573A (zh) * 2023-07-25 2023-08-22 讯飞医疗科技股份有限公司 一种处方用药审核方法、装置、设备及存储介质

Citations (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107731270A (zh) * 2017-10-25 2018-02-23 康美药业股份有限公司 一种自动审核药方准确性的方法及装置
CN108899070A (zh) * 2018-05-31 2018-11-27 平安医疗科技有限公司 处方推荐生成方法、装置、计算机设备和存储介质
CN109920508A (zh) * 2018-12-28 2019-06-21 安徽省立医院 处方审核方法及系统
CN110853723A (zh) * 2019-11-24 2020-02-28 曹庆恒 一种智能药品指导的方法、系统和设备
WO2020226456A1 (ko) * 2019-05-09 2020-11-12 박회준 처방 정보를 통한 의료 정보 제공 방법 및 장치
JP6890364B1 (ja) * 2021-02-19 2021-06-18 株式会社ミナケア 処方支援システム、処方支援方法、情報処理装置およびそれらの方法およびプログラム
CN113764066A (zh) * 2021-03-29 2021-12-07 北京京东拓先科技有限公司 一种处方数据的审核方法和装置
CN113782146A (zh) * 2021-09-16 2021-12-10 平安科技(深圳)有限公司 基于人工智能的全科用药推荐方法、装置、设备及介质
CN114550870A (zh) * 2022-03-23 2022-05-27 康键信息技术(深圳)有限公司 基于人工智能的处方审核方法、装置、设备及介质

Patent Citations (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107731270A (zh) * 2017-10-25 2018-02-23 康美药业股份有限公司 一种自动审核药方准确性的方法及装置
CN108899070A (zh) * 2018-05-31 2018-11-27 平安医疗科技有限公司 处方推荐生成方法、装置、计算机设备和存储介质
CN109920508A (zh) * 2018-12-28 2019-06-21 安徽省立医院 处方审核方法及系统
WO2020226456A1 (ko) * 2019-05-09 2020-11-12 박회준 처방 정보를 통한 의료 정보 제공 방법 및 장치
CN110853723A (zh) * 2019-11-24 2020-02-28 曹庆恒 一种智能药品指导的方法、系统和设备
JP6890364B1 (ja) * 2021-02-19 2021-06-18 株式会社ミナケア 処方支援システム、処方支援方法、情報処理装置およびそれらの方法およびプログラム
CN113764066A (zh) * 2021-03-29 2021-12-07 北京京东拓先科技有限公司 一种处方数据的审核方法和装置
CN113782146A (zh) * 2021-09-16 2021-12-10 平安科技(深圳)有限公司 基于人工智能的全科用药推荐方法、装置、设备及介质
CN114550870A (zh) * 2022-03-23 2022-05-27 康键信息技术(深圳)有限公司 基于人工智能的处方审核方法、装置、设备及介质

Also Published As

Publication number Publication date
CN114550870A (zh) 2022-05-27

Similar Documents

Publication Publication Date Title
US11232365B2 (en) Digital assistant platform
WO2023178978A1 (zh) 基于人工智能的处方审核方法、装置、设备及介质
US20080104104A1 (en) Health integration platform schema
US20190384849A1 (en) Data platform for automated data extraction, transformation, and/or loading
WO2022222943A1 (zh) 科室推荐方法、装置、电子设备及存储介质
CN113674858B (zh) 在线医疗处方用药智能检查方法、装置、设备及存储介质
CN113836131B (zh) 一种大数据清洗方法、装置、计算机设备及存储介质
WO2022194062A1 (zh) 疾病标签检测方法、装置、电子设备及存储介质
CN115238670B (zh) 信息文本抽取方法、装置、设备及存储介质
CN113887941A (zh) 业务流程生成方法、装置、电子设备及介质
US10936962B1 (en) Methods and systems for confirming an advisory interaction with an artificial intelligence platform
WO2022227171A1 (zh) 关键信息提取方法、装置、电子设备及介质
CN114840684A (zh) 基于医疗实体的图谱构建方法、装置、设备及存储介质
CN111403011B (zh) 挂号科室推送方法、装置、系统、电子设备及存储介质
WO2023178979A1 (zh) 问题标注方法、装置、电子设备及存储介质
CN116578704A (zh) 文本情感分类方法、装置、设备及计算机可读介质
CN116861875A (zh) 基于人工智能的文本处理方法、装置、设备及存储介质
CN116719904A (zh) 基于图文结合的信息查询方法、装置、设备及存储介质
CN116705304A (zh) 基于图像文本的多模态任务处理方法、装置、设备及介质
CN116702776A (zh) 基于跨中西医的多任务语义划分方法、装置、设备及介质
CN113689924B (zh) 相似病历检索方法、装置、电子设备及可读存储介质
CN115982454A (zh) 基于用户画像的问卷推送方法、装置、设备及存储介质
CN114676307A (zh) 基于用户检索的排序模型训练方法、装置、设备及介质
US20210133627A1 (en) Methods and systems for confirming an advisory interaction with an artificial intelligence platform
CN112507230B (zh) 基于浏览器的网页推荐方法、装置、电子设备及存储介质

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: 22933030

Country of ref document: EP

Kind code of ref document: A1