WO2023029510A1 - 基于人工智能的远程问诊方法、装置、设备及介质 - Google Patents

基于人工智能的远程问诊方法、装置、设备及介质 Download PDF

Info

Publication number
WO2023029510A1
WO2023029510A1 PCT/CN2022/087815 CN2022087815W WO2023029510A1 WO 2023029510 A1 WO2023029510 A1 WO 2023029510A1 CN 2022087815 W CN2022087815 W CN 2022087815W WO 2023029510 A1 WO2023029510 A1 WO 2023029510A1
Authority
WO
WIPO (PCT)
Prior art keywords
user
medical
information
preset
consultation
Prior art date
Application number
PCT/CN2022/087815
Other languages
English (en)
French (fr)
Inventor
宋泽友
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 WO2023029510A1 publication Critical patent/WO2023029510A1/zh

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
    • G16H50/00—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics
    • G16H50/20—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for computer-aided diagnosis, e.g. based on medical expert systems
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
    • G06F16/33—Querying
    • G06F16/3331—Query processing
    • G06F16/334—Query execution
    • G06F16/3344—Query execution using natural language analysis
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
    • G06F16/33—Querying
    • G06F16/3331—Query processing
    • G06F16/334—Query execution
    • G06F16/3346—Query execution using probabilistic model
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
    • G06F16/35—Clustering; Classification
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/24—Classification techniques
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F40/00—Handling natural language data
    • G06F40/20—Natural language analysis
    • G06F40/205—Parsing
    • G06F40/216—Parsing using statistical methods
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F40/00—Handling natural language data
    • G06F40/20—Natural language analysis
    • G06F40/279—Recognition of textual entities
    • G06F40/289—Phrasal analysis, e.g. finite state techniques or chunking
    • 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
    • G16H80/00—ICT specially adapted for facilitating communication between medical practitioners or patients, e.g. for collaborative diagnosis, therapy or health monitoring

Definitions

  • the present application relates to the technical field of artificial intelligence, and in particular to an artificial intelligence-based remote consultation method, device, equipment and storage medium.
  • Remote assisted consultation is only manually analyzed based on the patient's verbal description of somatosensory, or only using rule sentences to match and analyze the single-dimensional information of somatosensory description, without intelligently analyzing the rest of the patient's information (such as physical sign data) Analysis, resulting in the inability to accurately analyze the patient's consultation intention, so that the appropriate department or doctor cannot be accurately assigned to the patient.
  • This application provides an artificial intelligence-based remote consultation method, which includes:
  • the index information used to indicate the physical sign data in the medical inquiry information is input into the pre-established index recognition model to obtain the index recognition of the medical inquiry information result;
  • the present application also provides a remote medical consultation device based on artificial intelligence, which includes:
  • Judging module used to obtain the medical inquiry information uploaded by the user using the preset terminal, and judge the user type to which the user belongs based on the user's identification;
  • the first processing module when it is judged that the user belongs to the user who conducts medical consultation based on the medical consultation text, extracting the keywords of the medical consultation text in the medical consultation information, and feeding back candidate departments to the user based on the keywords option, the receiving user feeds back the consultation information to the terminal corresponding to the department based on the option of the candidate department option;
  • the second processing module when it is judged that the user belongs to the user who conducts medical consultation based on the physical sign data and images, input the index information used to indicate the physical sign data in the medical inquiry information into the pre-established index recognition model to obtain the obtained The index identification results of the medical inquiry information;
  • Sending module used to extract the characteristic information of the image in the medical inquiry information, input the characteristic information into the pre-established image recognition model, obtain the image recognition result of the medical inquiry information, and transfer the medical inquiry information and index recognition results and the image recognition result are sent to the terminal corresponding to the type of the indicator recognition result and the type of the image recognition result.
  • the present application also provides an electronic device, the electronic device comprising:
  • the memory stores a program that can be executed by the at least one processor, and the program is executed by the at least one processor, so that the at least one processor can perform the following remote consultation based on artificial intelligence Any step of the method:
  • the index information used to indicate the physical sign data in the medical inquiry information is input into the pre-established index recognition model to obtain the index recognition of the medical inquiry information result;
  • the present application also provides a computer-readable storage medium, the computer-readable storage medium stores a remote consultation program based on artificial intelligence, and when the remote consultation program based on artificial intelligence is executed by a processor, the following Arbitrary steps of the remote consultation method based on artificial intelligence:
  • the index information used to indicate the physical sign data in the medical inquiry information is input into the pre-established index recognition model to obtain the index recognition of the medical inquiry information result;
  • Fig. 1 is the schematic flow chart diagram of the preferred embodiment of the remote consultation method based on artificial intelligence of the present application
  • Fig. 2 is the module schematic diagram of the preferred embodiment of the remote medical interrogation device based on artificial intelligence of the present application;
  • FIG. 3 is a schematic diagram of a preferred embodiment of the electronic device of the present application.
  • AI artificial intelligence
  • the embodiments of the present application may acquire and process relevant data based on artificial intelligence technology.
  • artificial intelligence is the theory, method, technology and application system that uses 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. .
  • Artificial intelligence basic technologies generally include technologies such as sensors, dedicated artificial intelligence chips, cloud computing, distributed storage, big data processing technology, operation/interaction systems, and mechatronics.
  • Artificial intelligence software technology mainly includes computer vision technology, robotics technology, biometrics technology, speech processing technology, natural language processing technology, and machine learning/deep learning.
  • FIG. 1 it is a schematic flow chart of an embodiment of an artificial intelligence-based remote consultation method of the present application.
  • the method can be executed by an electronic device, and the electronic device can be realized by software and/or hardware.
  • Artificial intelligence-based remote consultation methods include:
  • Step S10 Obtain the medical inquiry information uploaded by the user using the preset terminal, and determine the user type to which the user belongs based on the user's identifier.
  • the user when the user needs to consult a doctor online, the user can start the application for online consultation on the terminal (such as a mobile phone or a smart watch, etc.) and initiate a request for consultation, and follow the prompts of the application Upload relevant medical inquiry information, wherein the medical inquiry information can include text information, index information, image information and/or video information, after the user uploads the medical inquiry information, he can submit the medical inquiry information and initiate a medical inquiry request, the server receives the terminal After the medical inquiry request is issued, the request is parsed to obtain the medical inquiry information carried in the request, wherein the request may include relevant medical inquiry information or a storage path of pending relevant medical inquiry information. That is to say, the medical consultation information may be input by the user when submitting the medical consultation request, or may be obtained by the application program from the address specified in the request after the user submits the medical consultation request.
  • the medical consultation information may be input by the user when submitting the medical consultation request, or may be obtained by the application program from the address specified in the request after the user submits the medical
  • the request also includes the user's identification (for example, ID number), which type of user the user can be judged according to the user's identification.
  • the user type includes the first type of user, the second type of user, and the third type of user.
  • the first type of user can be an ordinary patient who has registered an account (this type of user is a user who conducts consultation based on the consultation text), and the second type of user can be a postoperative recovery patient with a registered account or a chronic disease at home. Patients (this type of user is a user who conducts consultation based on physical sign data and images), and the third type of user can be an unregistered account consultation patient.
  • the type of the user can be judged through the user's identification, and then relevant remote consultation can be carried out according to the user type.
  • the method further includes:
  • a preset registration interface is fed back to the user, so that the user can input and upload the basic information of the user based on the registration interface for registration.
  • Unregistered users need to register an account before they can conduct remote medical consultation. If the user enters the consultation information for the first time in the online consultation application, the system will prompt the user to register first. After the registration is successful, the user needs to register in the system. After establishing health records and improving personal information, users can upload videos and images to the system through terminals or special wearable medical devices, and the system will conduct preliminary analysis and diagnosis.
  • Step S20 When it is judged that the user belongs to the user who conducts medical consultation based on the medical consultation text, extract the keywords of the medical consultation text in the medical consultation information, and feed back candidate department options to the user based on the keywords, and receive user Based on the option of the candidate department option, the consultation information is fed back to the terminal corresponding to the department.
  • the user's medical inquiry information may contain text information that has nothing to do with the content of the medical inquiry, it is necessary to obtain information from the medical inquiry information input by the user.
  • Extract key information When looking for the corresponding candidate department for the user based on the key information, the keyword extraction algorithm can be used to extract the key text information in the medical inquiry information.
  • the extracted keywords "eyelid” and "itch” assign the ophthalmology department and the dermatology department to the user as candidate departments, receive the user's options based on the candidate department options, and send the consultation information to the terminal corresponding to the department.
  • the user can choose a doctor or the system randomly assigns the corresponding doctor for diagnosis, and the doctor can issue a relevant treatment plan after communicating with the user.
  • said extracting keywords of medical questioning text in said medical questioning information includes:
  • Calculate the word frequency of each participle in the interrogation text calculate the IDF value and TF value of each participle based on the word frequency, multiply the IDF value of each participle with the TF value corresponding to each participle to obtain the TF-IDF of each participle value;
  • a first preset number of words with preset parts of speech are selected as the keywords based on the TF-IDF value of each word segment.
  • the interrogation text in the interrogation information After obtaining the interrogation text in the interrogation information, count the occurrence times of all words in the interrogation text, calculate the IDF (inverse document frequency value), and then calculate the TF (term frequency) of each word in the interrogation text value.
  • TF (the number of times the word appears in the text)/(the sum of the number of times each word appears in the text)
  • multiply the IDF value and the TF value to get the TF-IDF value of the word
  • the TF-IDF value can evaluate the word The importance of the word to the text, the larger the TF-IDF value, the higher the priority as a keyword.
  • the TF-IDF value of a word is obtained by comparing word frequency and inverse document frequency.
  • TF-IDF If the TF-IDF value is larger, the word is more important to the text, so TF-IDF can be used
  • the words with the highest IDF value are used as the keywords of the medical inquiry text, for example, the nouns with the top 3 TF-IDF values are selected as the keywords of the medical inquiry information text.
  • the feeding back candidate department options to the user based on the keywords includes:
  • the matching value may include a cosine similarity value.
  • Each department has a pre-established corpus table, which stores a large amount of medical inquiry text information corresponding to the department (that is, the sentence of the medical inquiry).
  • the medical questioning sentence can be crawled from the public website or entered manually It can also be automatically generated based on the BERT model.
  • the corpus of each department has a pre-labeled label set.
  • the label set of the ophthalmology corpus includes tags such as "eye”, “eye corner”, and “eyelid”
  • the label set of the dermatology corpus includes "itch”, "redness, and swelling”. "wait.
  • the department corresponding to the statement is used as a candidate department, for example, when the query text is matched with the sentences in the department corpus of the two label sets, there are sentences with a matching degree greater than 85%, then these two departments are used as candidate departments, And the candidate department corresponding to the sentence with the second highest matching value is marked with preset recommendation information.
  • the corpus of the ophthalmology department matches the inquiry text
  • the corpus of the dermatology department matches the inquiry text. If there is a statement with a matching degree of 85% during text matching, "recommendation" can be marked next to the words of ophthalmology, and feedback is given to the user, so that the user can choose the relevant department.
  • Step S30 When it is judged that the user belongs to the user who conducts medical consultation based on the physical sign data and images, input the index information used to indicate the physical sign data in the medical questioning information into the pre-established index recognition model to obtain the medical questioning information The index identification results.
  • the index information (such as body temperature, blood pressure, pulse, etc.) in the medical inquiry information can be obtained.
  • postoperative recovery patients and patients with some chronic diseases can wear wearable medical equipment during their home recuperation period, and upload relevant data such as user body temperature, blood pressure, heart rate, etc. to the remote consultation system as consultation information, or through terminals such as mobile phones Enter how you feel about your body (for example, whether you feel dizzy, weak in limbs, etc.).
  • the index information in the medical inquiry information is input into the pre-established index recognition model, and the index recognition result of the medical inquiry information is obtained.
  • the index identification result can be used as auxiliary information to help doctors make further diagnosis. It is also possible to synchronize the user's indicator information to the data model library as sample data for updating the indicator identification model.
  • the training process of the indicator recognition model includes:
  • Obtaining a third preset quantity of indicator data sets after performing preprocessing operations on each indicator data set, assigning preset labels to each indicator data set, using the indicator data set as an independent variable, and the corresponding preset label of the indicator data set as The dependent variable generates a training sample set;
  • the training is ended to obtain the indicator identification model.
  • the data model library can be established in advance.
  • the data model library stores a large number of user's sign data sets and corresponding disease types, and normalizes the user's vital sign data in the data model library to establish common diseases and user vital signs.
  • the correlation between data that is, the common diseases related to the user's vital sign data set, for example, the body temperature data corresponding to disease A is M, the blood pressure data is N, and so on.
  • the index data set is used as an independent variable, and the preset label corresponding to the index data set is used as a dependent variable to generate a training sample set.
  • the independent variable and the dependent variable can ensure that the model has excellent explanatory ability and predictive effect.
  • the training sample set is divided into a training set and a verification set according to a preset ratio (3:1), and the convolutional neural network model is trained using each variable and each dependent variable, and each variable and each dependent variable in the verification set is used every preset cycle
  • the accuracy rate of the convolutional neural network model is verified, and when the accuracy rate is greater than a second preset threshold (for example, 90%), the training is terminated to obtain an indicator identification model.
  • a second preset threshold for example, 90%
  • Step S40 Extract the characteristic information of the image in the medical inquiry information, input the characteristic information into the pre-established image recognition model, obtain the image recognition result of the medical inquiry information, and combine the medical inquiry information, index recognition results and image
  • the recognition result is sent to a terminal corresponding to the type of the indicator recognition result and the type of the image recognition result.
  • the feature information (such as texture features and color features) of the image information in the medical inquiry information can be extracted, and the feature information can be input into the pre-established image recognition model to obtain the image recognition result of the medical inquiry information, and the image recognition model can be established in advance.
  • the model can be obtained according to the convolutional neural network training, and the consultation information, index recognition results and image recognition results are sent to the terminal of the corresponding doctor. For example, if it is recognized that the image in the user’s consultation information is about swollen face, dark circles under the eyes, and yellow complexion, it may identify the user’s abnormal liver function, etc., so the user is assigned to a doctor related to liver function to complete the intelligent diagnosis system. Accurate diagnostic analysis.
  • Doctors can make medical diagnosis suggestions based on intelligent diagnosis information, patient medical records and patient’s self-generated body sensations, judge whether users should continue to observe at home or need to return to the hospital for treatment, and communicate with patients to give corresponding diagnosis and treatment plans, such as new prescriptions , increase exercise, dietary precautions, etc., so that doctors can guide patients' treatment and post-diagnosis recovery by analyzing cases, conditions, and patient-related vital signs at the remote end.
  • said extracting feature information of images in said medical inquiry information includes:
  • Using the LBP algorithm to extract the texture features of the image cutting the image into several cut images, using the color moment extraction algorithm to extract the color features of each cut image, performing fusion on the texture features and the color features of each cut image
  • the fusion feature of the image is obtained through the operation, and the fusion feature is used as feature information of the image.
  • the detection window is first divided into 16 ⁇ 16 regions (cells), and for a pixel in each cell, 8 points in its circular neighborhood (or multiple points in the circular neighborhood) are clockwise Or counterclockwise comparison, if the center pixel value is larger than the adjacent point, then assign the adjacent point a value of 1, otherwise assign a value of 0, so that each point will get an 8-bit binary number (usually converted to a decimal number), and then calculate The histogram of each cell, that is, the frequency of each number (assumed to be a decimal number) (that is, a binary sequence about whether each pixel is larger than the points in the neighborhood), and then normalize the histogram processing, and finally connect the obtained statistical histograms of each cell to obtain the LBP texture feature of the image.
  • the color moment extraction algorithm uses the concept of moments in linear algebra to express the color distribution in the image by its moments.
  • the color distribution is described by the first moment of color (average), second moment of color (variance) and third moment of color (skewness).
  • image description using color moments does not need to quantify image features. Since each pixel has three color channels of the color space, the color moment of the image has 9 components to describe. Since color moments have fewer dimensions, color moments can be used in combination with other image features.
  • the clipping of the image into several cropped images includes:
  • the coordinate origin is set at the lower left corner of the image, and the image is cropped along the x-axis and y-axis respectively with a preset step size to obtain several corresponding cropped images of a preset size.
  • cropped images corresponding to an image are classified into a cropped image set. For example, set the origin at the lower left corner of the image, and use the left boundary of the image as the Y axis, the lower boundary of the image as the X axis, and crop with a step size of 32 pixels to obtain several cropped images as the cropped image set of the image .
  • FIG. 2 it is a schematic diagram of functional modules of the artificial intelligence-based remote consultation device 100 of the present application.
  • the artificial intelligence-based remote consultation device 100 described in this application can be installed in electronic equipment. According to the realized functions, the artificial intelligence-based remote consultation device 100 may include a judging module 110 , a first processing module 120 , a second processing module 130 and a sending module 140 .
  • the module described in the present invention can also be referred to as 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 fixed functions, and are stored in the memory of the electronic device.
  • each module/unit is as follows:
  • Judging module 110 used to acquire the medical inquiry information uploaded by the user using the preset terminal, and judge the user type to which the user belongs based on the user's identifier.
  • the first processing module 120 is configured to extract keywords of the medical questioning text in the medical questioning information when it is judged that the user belongs to the user who conducts medical questioning based on the medical questioning text, and feed back candidates to the user based on the keywords Department options, the receiving user feeds back the consultation information to the terminal corresponding to the department based on the options of the candidate department options.
  • the second processing module 130 is configured to input the indicator information used to indicate the physical sign data in the medical inquiry information into the pre-established indicator recognition model when it is judged that the user belongs to the user who conducts medical consultation based on the physical sign data and images, and obtains The index identification result of the medical inquiry information.
  • the sending module 140 is used to extract the characteristic information of the image in the medical inquiry information, input the characteristic information into the pre-established image recognition model, obtain the image recognition result of the medical inquiry information, and identify the medical inquiry information and indicators The result and the image recognition result are sent to a terminal corresponding to the type of the indicator recognition result and the type of the image recognition result.
  • the judging module 110 is also used for:
  • a preset registration interface is fed back to the user, so that the user can input and upload the basic information of the user based on the registration interface for registration.
  • said extracting keywords of medical questioning text in said medical questioning information includes:
  • Calculate the word frequency of each participle in the interrogation text calculate the IDF value and TF value of each participle based on the word frequency, multiply the IDF value of each participle with the TF value corresponding to each participle to obtain the TF-IDF of each participle value;
  • a first preset number of words with preset parts of speech are selected as the keywords based on the TF-IDF value of each word segment.
  • the feeding back candidate department options to the user based on the keywords includes:
  • the department corresponding to the sentence is used as the candidate department, and the candidate department corresponding to the sentence with the highest sentence matching degree value is marked with preset recommendation information .
  • the training process of the indicator recognition model includes:
  • Obtaining a third preset quantity of indicator data sets after performing preprocessing operations on each indicator data set, assigning preset labels to each indicator data set, using the indicator data set as an independent variable, and the corresponding preset label of the indicator data set as The dependent variable generates a training sample set;
  • the training is ended to obtain the indicator identification model.
  • said extracting feature information of images in said medical inquiry information includes:
  • Using the LBP algorithm to extract the texture features of the image cutting the image into several cut images, using the color moment extraction algorithm to extract the color features of each cut image, performing fusion on the texture features and the color features of each cut image
  • the fusion feature of the image is obtained through the operation, and the fusion feature is used as feature information of the image.
  • the cropping the image into several cropped images includes:
  • the coordinate origin is set at the lower left corner of the image, and the image is cropped along the x-axis and y-axis respectively with a preset step size to obtain several corresponding cropped images of a preset size.
  • FIG. 3 it is a schematic diagram of a preferred embodiment of the electronic device 1 of the present application.
  • the electronic device 1 includes but not limited to: a memory 11 , a processor 12 , a display 13 and a network interface 14 .
  • the electronic device 1 is connected to the network through the network interface 14 to obtain raw data.
  • the network may be an intranet (Intranet), the Internet (Internet), the Global System for Mobile Communications (Global System of Mobile communication (GSM), Wideband Code Division Multiple Access (WCDMA), 4G network, 5G network, Bluetooth (Bluetooth), Wi-Fi, call network and other wireless or wired networks.
  • the memory 11 includes at least one type of readable storage medium
  • the readable storage medium includes flash memory, hard disk, multimedia card, card-type memory (for example, SD or DX memory, etc.), random access memory (RAM), static Random access memory (SRAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), programmable read-only memory (PROM), magnetic memory, magnetic disk, optical disk, etc.
  • the computer readable storage Media can be either non-volatile or volatile.
  • the storage 11 may be an internal storage unit of the electronic device 1 , such as a hard disk or a memory of the electronic device 1 .
  • the memory 11 can also be an external storage device of the electronic device 1, such as a plug-in hard disk equipped with the electronic device 1, a smart memory card (Smart Media Card, SMC), a secure digital ( secure Digital, SD) card, flash memory card (Flash Card), etc.
  • the storage 11 may also include both an internal storage unit of the electronic device 1 and an external storage device thereof.
  • the memory 11 is generally used to store the operating system and various application software installed in the electronic device 1 , such as program codes of the artificial intelligence-based remote consultation program 10 .
  • the memory 11 can also be used to temporarily store various types of data that have been output or will be output.
  • Processor 12 may be a central processing unit (Central Processing Unit) in some embodiments Processing Unit, CPU), controller, microcontroller, microprocessor, or other data processing chips.
  • the processor 12 is generally used to control the overall operation of the electronic device 1 , for example, perform data interaction or communication-related control and processing.
  • the processor 12 is configured to run program codes stored in the memory 11 or process data, for example, run program codes of the artificial intelligence-based remote consultation program 10 .
  • the display 13 may be called a display screen or a display unit.
  • the display 13 can be an LED display, a liquid crystal display, a touch-sensitive liquid crystal display, and an organic light-emitting diode (Organic Light-Emitting Diode, OLED) touch panel, etc.
  • the display 13 is used for displaying the information processed in the electronic device 1 and for displaying a visualized working interface, such as displaying the results of statistical data.
  • the network interface 14 may optionally include a standard wired interface or wireless interface (such as a WI-FI interface), and the network interface 14 is generally used to establish a communication connection between the electronic device 1 and other electronic devices.
  • a standard wired interface or wireless interface such as a WI-FI interface
  • Fig. 3 only shows the electronic device 1 with components 11-14 and the remote consultation program 10 based on artificial intelligence, but it should be understood that it is not required to implement all the components shown, and more or more components can be implemented instead. few components.
  • the electronic device 1 may further include a user interface, and the user interface may include a display (Display), an input unit such as a keyboard (Keyboard), and an optional user interface may also include a standard wired interface and a wireless interface.
  • the display may be an LED display, a liquid crystal display, a touch-sensitive liquid crystal display, an organic light-emitting diode (Organic Light-Emitting Diode, OLED) touch device, and the like.
  • the display may also be appropriately called a display screen or a display unit, and is used for displaying information processed in the electronic device 1 and for displaying a visualized user interface.
  • the electronic device 1 may also include a radio frequency (Radio Frequency, RF) circuit, a sensor, an audio circuit, etc., which will not be repeated here.
  • RF Radio Frequency
  • the index information used to indicate the physical sign data in the medical inquiry information is input into the pre-established index recognition model to obtain the index recognition of the medical inquiry information result;
  • the storage device may be the memory 11 of the electronic device 1 , or other storage devices connected in communication with the electronic device 1 .
  • the embodiment of the present application also provides a computer-readable storage medium, and the computer-readable storage medium may be non-volatile or volatile.
  • the computer-readable storage medium can be hard disk, multimedia card, SD card, flash memory card, SMC, read-only memory (ROM), erasable programmable read-only memory (EPROM), portable compact disk read-only memory (CD- ROM), USB memory, etc., or any combination of several.
  • the computer-readable storage medium includes a storage data area and a storage program area, the storage data area stores data created according to the use of blockchain nodes, and the storage program area stores a remote consultation program 10 based on artificial intelligence. Realize the following operations when the remote consultation program 10 based on artificial intelligence is executed by the processor:
  • the index information used to indicate the physical sign data in the medical inquiry information is input into the pre-established index recognition model to obtain the index recognition of the medical inquiry information result;
  • all the above-mentioned data can also be stored in a block chain node.
  • the results of indicator recognition and image recognition, etc. these data can be stored in the blockchain nodes.
  • Blockchain essentially a decentralized database, is a series of data blocks associated with each other using cryptographic methods. Each data block contains a batch of network transaction information, which is used to verify its Validity of information (anti-counterfeiting) and generation of the next block.
  • the blockchain can include the underlying platform of the blockchain, the platform product service layer, and the application service layer.
  • the methods of the above embodiments can be implemented by means of software plus a necessary general-purpose hardware platform, and of course also by hardware, but in many cases the former is better implementation.
  • the technical solution of the present application can be embodied in the form of a software product in essence or the part that contributes to the prior art, and the computer software product is stored in a storage medium as described above (such as ROM/RAM , magnetic disk, optical disk), including several instructions to enable a terminal device (which may be a mobile phone, computer, electronic device, or network device, etc.) to execute the methods described in various embodiments of the present application.

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Health & Medical Sciences (AREA)
  • Data Mining & Analysis (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Databases & Information Systems (AREA)
  • General Health & Medical Sciences (AREA)
  • Computational Linguistics (AREA)
  • Medical Informatics (AREA)
  • Artificial Intelligence (AREA)
  • Biomedical Technology (AREA)
  • Public Health (AREA)
  • Pathology (AREA)
  • Epidemiology (AREA)
  • Primary Health Care (AREA)
  • Probability & Statistics with Applications (AREA)
  • Audiology, Speech & Language Pathology (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Evolutionary Biology (AREA)
  • Evolutionary Computation (AREA)
  • Medical Treatment And Welfare Office Work (AREA)
  • Information Retrieval, Db Structures And Fs Structures Therefor (AREA)

Abstract

本申请涉及人工智能技术领域，提供了一种基于人工智能的远程问诊方法、装置、设备及存储介质。所述方法包括：获取用户上传的问诊信息并判断用户的类型，若为普通患者，基于问诊信息中问诊文本的关键词向用户反馈候选科室选项，接收用户基于候选科室的选项，将问诊信息反馈至该科室对应的终端，当判断用户为术后患者时，将问诊信息中的指标信息输入指标识别模型得到指标识别结果，将问诊信息中图像的特征信息输入图像识别模型得到图像识别结果，将问诊信息、指标识别结果及图像识别结果发送至识别结果类型对应的终端。本申请可以提高远程问诊辅助的准确度。本申请还涉及区块链技术领域，上述识别结果可以存储于一区块链的节点中。

Description

基于人工智能的远程问诊方法、装置、设备及介质
本申请要求于2021年11月26日提交中国专利局、申请号为202111003077.0，发明名称为“基于人工智能的远程问诊方法、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远程问诊方法、装置、设备及存储介质。
 背景技术
随着人工智能的高速发展以及医院信息化的发展，智能远程辅助问诊方式也随之产生。发明人意识到目前由于大多数患者的自身医疗知识有限，多数患者很难准确地判断自己要挂号或者问诊咨询的科室，此外，对于手术后恢复期患者或在家疗养的慢性病患者，现有的远程辅助问诊也仅是依据患者对体感的口头描述人工进行分析，或者仅利用规则语句对体感描述这单一维度的信息进行匹配分析，而未对患者的其余信息（如，体征数据）进行智能分析，导致不能准确分析出患者的问诊意图，从而不能准确地给患者分配合适的科室或医生。
 技术解决方案
本申请提供一种基于人工智能的远程问诊方法，该方法包括：
获取用户利用预设终端上传的问诊信息，基于所述用户的标识判断所述用户所属的用户类型；
当判断所述用户属于基于问诊文本进行问诊的用户时，提取所述问诊信息中问诊文本的关键词，基于所述关键词向所述用户反馈候选科室选项，接收用户基于所述候选科室选项的选项将所述问诊信息反馈至该科室对应的终端；
当判断所述用户属于基于体征数据和图像进行问诊的用户时，将所述问诊信息中用于指示体征数据的指标信息输入预先建立的指标识别模型，得到所述问诊信息的指标识别结果；
提取所述问诊信息中图像的特征信息，将所述特征信息输入预先建立图像识别模型，得到所述问诊信息的图像识别结果，将所述问诊信息、指标识别结果及图像识别结果，发送至所述指标识别结果的类型及所述图像识别结果的类型对应的终端。
本申请还提供一种基于人工智能的远程问诊装置，该基于人工智能的远程问诊装置包括：
判断模块：用于获取用户利用预设终端上传的问诊信息，基于所述用户的标识判断所述用户所属的用户类型；
第一处理模块：用于当判断所述用户属于基于问诊文本进行问诊的用户时，提取所述问诊信息中问诊文本的关键词，基于所述关键词向所述用户反馈候选科室选项，接收用户基于所述候选科室选项的选项将所述问诊信息反馈至该科室对应的终端；
第二处理模块：用于当判断所述用户属于基于体征数据和图像进行问诊的用户时，将所述问诊信息中用于指示体征数据的指标信息输入预先建立的指标识别模型，得到所述问诊信息的指标识别结果；
发送模块：用于提取所述问诊信息中图像的特征信息，将所述特征信息输入预先建立图像识别模型，得到所述问诊信息的图像识别结果，将所述问诊信息、指标识别结果及图像识别结果，发送至所述指标识别结果的类型及所述图像识别结果的类型对应的终端。
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程序，所述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远程问诊方法的任意步骤：
获取用户利用预设终端上传的问诊信息，基于所述用户的标识判断所述用户所属的用户类型；
当判断所述用户属于基于问诊文本进行问诊的用户时，提取所述问诊信息中问诊文本的关键词，基于所述关键词向所述用户反馈候选科室选项，接收用户基于所述候选科室选项的选项将所述问诊信息反馈至该科室对应的终端；
当判断所述用户属于基于体征数据和图像进行问诊的用户时，将所述问诊信息中用于指示体征数据的指标信息输入预先建立的指标识别模型，得到所述问诊信息的指标识别结果；
提取所述问诊信息中图像的特征信息，将所述特征信息输入预先建立图像识别模型，得到所述问诊信息的图像识别结果，将所述问诊信息、指标识别结果及图像识别结果，发送至所述指标识别结果的类型及所述图像识别结果的类型对应的终端。
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存储有基于人工智能的远程问诊程序，所述基于人工智能的远程问诊程序被处理器执行时，实现如下所述基于人工智能的远程问诊方法的任意步骤：
获取用户利用预设终端上传的问诊信息，基于所述用户的标识判断所述用户所属的用户类型；
当判断所述用户属于基于问诊文本进行问诊的用户时，提取所述问诊信息中问诊文本的关键词，基于所述关键词向所述用户反馈候选科室选项，接收用户基于所述候选科室选项的选项将所述问诊信息反馈至该科室对应的终端；
当判断所述用户属于基于体征数据和图像进行问诊的用户时，将所述问诊信息中用于指示体征数据的指标信息输入预先建立的指标识别模型，得到所述问诊信息的指标识别结果；
提取所述问诊信息中图像的特征信息，将所述特征信息输入预先建立图像识别模型，得到所述问诊信息的图像识别结果，将所述问诊信息、指标识别结果及图像识别结果，发送至所述指标识别结果的类型及所述图像识别结果的类型对应的终端。
 附图说明
图1为本申请基于人工智能的远程问诊方法较佳实施例的流程图示意图；
图2为本申请基于人工智能的远程问诊装置较佳实施例的模块示意图；
图3为本申请电子设备较佳实施例的示意图；
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
为了使本申请的目的、技术方案及优点更加清楚明白，以下结合附图及实施例，对本申请进行进一步详细说明。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。基于本申请中的实施例，本领域普通技术人员在没有做出创造性劳动前提下所获得的所有其他实施例，都属于本申请保护的范围。
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（Artificial Intelligence，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。
人工智能基础技术一般包括如传感器、专用人工智能芯片、云计算、分布式存储、大数据处理技术、操作/交互系统、机电一体化等技术。人工智能软件技术主要包括计算机视觉技术、机器人技术、生物识别技术、语音处理技术、自然语言处理技术以及机器学习/深度学习等几大方向。
本申请提供一种基于人工智能的远程问诊方法。参照图1所示，为本申请基于人工智能的远程问诊方法的实施例的方法流程示意图。该方法可以由一个电子设备执行，该电子设备可以由软件和/或硬件实现。基于人工智能的远程问诊方法包括：
步骤S10：获取用户利用预设终端上传的问诊信息，基于所述用户的标识判断所述用户所属的用户类型。
在本实施例中，当用户需要在线上问诊时，用户可以开启终端（例如，手机或智能手表等）上用于线上问诊的应用程序并发起问诊请求，并根据应用程序的提示上传相关的问诊信息，其中，问诊信息可以包括文本信息、指标信息、图像信息及/或视频信息，用户上传问诊信息后，可以提交问诊信息并发起问诊请求，服务器接收到终端发出的问诊请求后，解析该请求获取请求中携带的问诊信息，其中，请求中可以包括相关的问诊信息，也可以包括待相关问诊信息的存储路径。也就是说，问诊信息可以是用户在提交问诊请求时一并输入，也可以是用户提交问诊请求之后应用程序从请求指定的地址中获取的。
请求中还包括用户的标识（例如，ID号），根据用户的标识可以判断用户为哪种类型的用户，本实施例中用户类型包括第一类型用户、第二类型用户、第三类型用户，其中，第一类型用户可以是已注册账号的普通患者（该类用户是基于问诊文本进行问诊的用户），第二类型用户可以是已注册账号的术后恢复期患者或在家疗养的慢性病患者（该类用户是基于体征数据和图像进行问诊的用户），第三类型用户可以是未注册账号的问诊患者。通过用户的标识可以判断用户所属的类型，进而根据用户类型进行相关的远程问诊。
在一个实施例中，在基于所述用户的标识判断所述用户所属的用户类型之后，所述方法还包括：
当判断所述用户属于未注册账号的用户时，将预设的注册界面反馈至所述用户，以供所述用户基于所述注册界面输入并上传所述用户的基本信息进行注册。
未注册的用户需要先注册账号才可以进行远程问诊，若用户第一次在线上问诊的应用程序输入问诊信息，则系统会提示该用户先进行注册，用户在注册成功后需要在系统建立健康档案并完善个人信息，之后，用户可将视频、图像通过终端或者特殊的可佩戴医疗设备上传至系统，由系统进行初步分析诊断。
步骤S20：当判断所述用户属于基于问诊文本进行问诊的用户时，提取所述问诊信息中问诊文本的关键词，基于所述关键词向所述用户反馈候选科室选项，接收用户基于所述候选科室选项的选项将所述问诊信息反馈至该科室对应的终端。
在本实施例中，若根据用户的标识判断出用户为已注册账号的普通患者，由于用户在问诊信息中可能包含与问诊内容无关的文本信息，因此需要从用户输入的问诊信息中提取出关键信息。根据关键信息为用户寻找对应的候选科室时，可以利用关键词提取算法提取出问诊信息中关键文本信息，例如，用户问诊信息中文本信息为“眼皮痒了一段时间了”，则可以根据提取出的关键词“眼皮”、“痒”向用户分配眼科科室和皮肤科科室作为候选科室，接收用户基于候选科室选项的选项，将问诊信息发送至该科室对应的终端。用户选择可以选择医生或者系统随机分配相应的医生进行诊断，医生和用户沟通后出可出具相关的治疗方案。
在一个实施例中，所述提取所述问诊信息中问诊文本的关键词，包括：
获取所述问诊信息中的问诊文本，并对所述问诊文本执行分词操作得到多个分词；
计算各分词在所述问诊文本中的词频，基于所述词频计算出各分词的IDF值及TF值，将各分词的IDF值与各分词对应的TF值相乘得到各分词的TF-IDF值；
基于各分词的TF-IDF值选取第一预设数量的预设词性的词作为所述关键词。
获取所述问诊信息中的问诊文本后，统计问诊文本中的所有词的出现次数，计算出IDF（逆文档频率值），再计算出问诊文本中每个词的TF（词频）值。其中，TF=（词语在文本中出现次数）/（各词语在文本中出现次数的总和），将IDF值与TF值相乘，得到该词的TF-IDF值，TF-IDF值可以评估字词对于文本中的重要程度，TF-IDF值越大表示作为关键词的优先级越高。在进行TF-IDF计算时，通过对词频与逆文档频率得出某个字词的TF-IDF值，若TF-IDF值越大，该词对文本的重要性越高，因此可以将TF-IDF值排在前面的字词作为问诊文本的关键词，例如，选取TF- IDF值排在前3的名词作为问诊信息文本的关键词。
在一个实施例中，所述基于所述关键词向所述用户反馈候选科室选项，包括：
将所述关键词分别与预设的科室语料表对应的标签集进行匹配得到多个标签集匹配度值，选取出标签集匹配度值最高的第二预设数量的候选标签集，将所述问诊文本分别与所述候选标签集对应科室语料表中的各语句进行匹配得到多个语句匹配度值；
当科室语料表中存在语句匹配度值大于第一预设阈值的语句时，将该语句对应的科室作为所述候选科室，并将语句匹配度值最高的语句对应的候选科室标注预设推荐信息。其中，匹配度值可以是包括余弦相似度值。
每个科室有预先建立的语料表，语料表中存储有大量的该科室对应的问诊文本信息（即问诊的语句），问诊语句可以是从公开网站爬取得到，也可以是人工录入的，还可以是基于BERT模型自动生成的。每个科室的语料表有预先标注的标签集，例如，眼科语料表的标签集中有“眼睛”、“眼角”、“眼皮”等标签，皮肤科语料表的标签集中有“痒”、“红肿”等。
将问诊文本中的关键词分别与科室语料表对应的标签集进行匹配，得到多个标签集匹配度值，选取出标签集匹配度值最高的两个或者三个候选标签集，将问诊文本分别与候选标签集对应科室语料表中的各语句进行匹配，得到多个第二匹配值，当科室语料表中存在第二匹配值大于第一预设阈值（例如，85%）的语句时，将该语句对应的科室作为候选科室，例如，问诊文本与两个标签集的科室语料表的句子匹配时均存在匹配度值大于85%的句子，则将这两个科室作为候选科室，并将第二匹配值最高的语句对应的候选科室标注预设推荐信息，例如，眼科科室的语料表与问诊文本匹配时存在匹配度值95%的语句，皮肤科科室的语料表与问诊文本匹配时存在匹配度值85%的语句，则可以在眼科的字眼旁边标注“推荐”，并反馈给用户，以供用户选择相关的科室。
步骤S30：当判断所述用户属于基于体征数据和图像进行问诊的用户时，将所述问诊信息中用于指示体征数据的指标信息输入预先建立的指标识别模型，得到所述问诊信息的指标识别结果。
在本实施例中，若根据用户的标识判断出用户为已注册账号的术后恢复期患者或在家疗养的慢性病患者，可以获取问诊信息中的指标信息（例如，体温、血压、脉搏等），术后恢复期患者和一些慢性病患者在居家疗养期间可以佩戴可佩戴式的医疗设备，将用户体温、血压、心率等相关数据作为问诊信息上传至远程问诊系统，也可以通过手机等终端输入自身的身体感觉情况（例如，有无头昏眼花、四肢无力等）。
获取用户的指标信息之后，将问诊信息中的指标信息输入预先建立的指标识别模型，得到问诊信息的指标识别结果。该指标识别结果可以作为辅助信息帮助医生做进一步的诊断。还可以将用户的指标信息同步至数据模型库，作为更新指标识别模型的样本数据。
在一个实施例中，所述指标识别模型的训练过程包括：
获取第三预设数量的指标数据集，对各指标数据集执行预处理操作后，分别为各指标数据集分配预设标签，将指标数据集作为自变量、指标数据集对应的预设标签作为因变量生成训练样本集；
将所述训练样本集按照预设比例分成训练集及验证集；
利用所述训练集中的各自变量及各因变量对卷积神经网络模型进行训练，每隔预设周期利用所述验证集中各变量及各因变量对卷积神经网络模型的准确率进行验证；及
当验证所述准确率大于第二预设阈值时，结束训练得到所述指标识别模型。
可以预先建立数据模型库，数据模型库存储有大量的用户的体征数据集及对应的疾病类型，将数据模型库中用户的各项生命体征数据作归一化处理，建立常见疾病和用户生命体征数据间的关联性，即用户生命体征数据集标注相关的常见疾病，例如，疾病A对应的体温数据是M，血压数据是N等等。将指标数据集作为自变量、指标数据集对应的预设标签作为因变量生成训练样本集，自变量和因变量可以保证模型具有优良的解释能力和预测效果。将训练样本集按照预设比例（3:1）分成训练集及验证集，利用各自变量及各因变量对卷积神经网络模型进行训练，每隔预设周期利用验证集中各变量及各因变量对卷积神经网络模型的准确率进行验证，当准确率大于第二预设阈值（例如，90%）时，结束训练得到指标识别模型。
步骤S40：提取所述问诊信息中图像的特征信息，将所述特征信息输入预先建立图像识别模型，得到所述问诊信息的图像识别结果，将所述问诊信息、指标识别结果及图像识别结果，发送至所述指标识别结果的类型及所述图像识别结果的类型对应的终端。
在本实施例中，可以提取问诊信息中图像信息的特征信息（例如，纹理特征和颜色特征），将特征信息输入预先建立图像识别模型，得到问诊信息的图像识别结果，预先建立图像识别模型可以是根据卷积神经网络训练得到的，将问诊信息、指标识别结果及图像识别结果，发送至对应医生的终端。例如，识别出用户问诊信息中的图像是关于脸部浮肿、眼圈发黑、脸色发黄，则可能识别出用户肝功能异常等，于是将用户分配给肝功能相关医生，完成智能诊断系统的精准诊断分析。
医生可以结合智能诊断信息，结合患者病历和患者自生的身体感觉做出医疗诊断建议，判断用户是继续居家观察疗养还是需要回医院治疗，并和患者沟通给出相应的诊治方案，如新开药方、增加锻炼、饮食注意事项等，实现医生在远端通过分析病例、病情及患者相关生命体征来指导患者治疗和诊后的恢复。
在一个实施例中，所述提取所述问诊信息中图像的特征信息，包括：
利用LBP算法提取出所述图像的纹理特征，将所述图像裁剪成若干张裁剪图像，利用颜色矩提取算法提取出各裁剪图像的颜色特征，对所述纹理特征及各裁剪图像颜色特征执行融合操作得到所述图像的融合特征，将所述融合特征作为所述图像的特征信息。
具体地，首先将检测窗口划分为16×16的区域（cell），对于每个cell中的一个像素，将其环形邻域内的8个点（也可以是环形邻域多个点）进行顺时针或逆时针的比较，如果中心像素值比该邻点大，则将邻点赋值为1，否则赋值为0，这样每个点都会获得一个8位二进制数（通常转换为十进制数），然后计算每个cell的直方图，即每个数字（假定是十进制数）出现的频率（即一个关于每一个像素点是否比邻域内点大的一个二进制序列进行统计），然后对该直方图进行归一化处理，最后将得到的每个cell的统计直方图进行连接，得到图像的LBP纹理特征。
颜色矩提取算法利用线性代数中矩的概念，将图像中的颜色分布用其矩表示。利用颜色一阶矩（平均值）、颜色二阶矩（方差）和颜色三阶矩（偏斜度）来描述颜色分布。其与颜色直方图不同，利用颜色矩进行图像描述无需量化图像特征。由于每个像素具有颜色空间的三个颜色通道，因此图像的颜色矩有9个分量来描述。由于颜色矩的维度较少，因此可以将颜色矩与其它的图像特征结合使用。
进一步地，所述将所述图像裁剪成若干张裁剪图像，包括：
在所述图像的左下角设置坐标原点，将所述图像分别沿着x轴和y轴方向，且以预设步长进行裁剪，得到若干张对应的预设大小的裁剪图像。
图像对应的若干张裁剪图像归为一个裁剪图像集合。例如，在图像的左下角设置原点，并以图像的左边界为Y轴、以图像的下边界为X轴，以步长为32像素进行裁剪，得到若干张裁剪图像作为该图像的裁剪图像集合。
参照图2所示，为本申请基于人工智能的远程问诊装置100的功能模块示意图。
本申请所述基于人工智能的远程问诊装置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远程问诊装置100可以包括判断模块110、第一处理模块120、第二处理模块130及发送模块140。本发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：
判断模块110：用于获取用户利用预设终端上传的问诊信息，基于所述用户的标识判断所述用户所属的用户类型。
第一处理模块120，用于当判断所述用户属于基于问诊文本进行问诊的用户时，提取所述问诊信息中问诊文本的关键词，基于所述关键词向所述用户反馈候选科室选项，接收用户基于所述候选科室选项的选项将所述问诊信息反馈至该科室对应的终端。
第二处理模块130，用于当判断所述用户属于基于体征数据和图像进行问诊的用户时，将所述问诊信息中用于指示体征数据的指标信息输入预先建立的指标识别模型，得到所述问诊信息的指标识别结果。
发送模块140，用于提取所述问诊信息中图像的特征信息，将所述特征信息输入预先建立图像识别模型，得到所述问诊信息的图像识别结果，将所述问诊信息、指标识别结果及图像识别结果，发送至所述指标识别结果的类型及所述图像识别结果的类型对应的终端。
在一个实施例中，判断模块110还用于：
当判断所述用户属于未注册账号的用户时，将预设的注册界面反馈至所述用户，以供所述用户基于所述注册界面输入并上传所述用户的基本信息进行注册。
在一个实施例中，所述提取所述问诊信息中问诊文本的关键词，包括：
获取所述问诊信息中的问诊文本，并对所述问诊文本执行分词操作得到多个分词；
计算各分词在所述问诊文本中的词频，基于所述词频计算出各分词的IDF值及TF值，将各分词的IDF值与各分词对应的TF值相乘得到各分词的TF-IDF值；
基于各分词的TF-IDF值选取第一预设数量的预设词性的词作为所述关键词。
在一个实施例中，所述基于所述关键词向所述用户反馈候选科室选项，包括：
将所述关键词分别与预设的科室语料表对应的标签集进行匹配得到多个标签集匹配度值，选取出标签集匹配度值最高的第二预设数量的候选标签集，将所述问诊文本分别与所述候选标签集对应科室语料表中的各语句进行匹配得到多个语句匹配度值；
当科室语料表中存在语句匹配度值大于第一预设阈值的语句时，将该语句对应的科室作为所述候选科室，并将语句匹配度值最高的语句对应的候选科室标注预设推荐信息。
在一个实施例中，所述指标识别模型的训练过程包括：
获取第三预设数量的指标数据集，对各指标数据集执行预处理操作后，分别为各指标数据集分配预设标签，将指标数据集作为自变量、指标数据集对应的预设标签作为因变量生成训练样本集；
将所述训练样本集按照预设比例分成训练集及验证集；
利用所述训练集中的各自变量及各因变量对卷积神经网络模型进行训练，每隔预设周期利用所述验证集中各变量及各因变量对卷积神经网络模型的准确率进行验证；及
当验证所述准确率大于第二预设阈值时，结束训练得到所述指标识别模型。
在一个实施例中，所述提取所述问诊信息中图像的特征信息，包括：
利用LBP算法提取出所述图像的纹理特征，将所述图像裁剪成若干张裁剪图像，利用颜色矩提取算法提取出各裁剪图像的颜色特征，对所述纹理特征及各裁剪图像颜色特征执行融合操作得到所述图像的融合特征，将所述融合特征作为所述图像的特征信息。
在一个实施例中，所述将所述图像裁剪成若干张裁剪图像，包括：
在所述图像的左下角设置坐标原点，将所述图像分别沿着x轴和y轴方向，且以预设步长进行裁剪，得到若干张对应的预设大小的裁剪图像。
参照图3所示，为本申请电子设备1较佳实施例的示意图。
该电子设备1包括但不限于：存储器11、处理器12、显示器13及网络接口14。所述电子设备1通过网络接口14连接网络，获取原始数据。其中，所述网络可以是企业内部网（Intranet）、互联网（Internet）、全球移动通讯系统（Global System of Mobile communication，GSM）、宽带码分多址（Wideband Code Division Multiple Access，WCDMA）、4G网络、5G网络、蓝牙（Bluetooth）、Wi-Fi、通话网络等无线或有线网络。
其中，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、硬盘、多媒体卡、卡型存储器（例如，SD或DX存储器等）、随机访问存储器（RAM）、静态随机访问存储器（SRAM）、只读存储器（ROM）、电可擦除可编程只读存储器（EEPROM）、可编程只读存储器（PROM）、磁性存储器、磁盘、光盘等，所述计算机可读存储介质可以是非易失性的，也可以是易失性的。在一些实施例中，所述存储器11可以是所述电子设备1的内部存储单元，例如该电子设备1的硬盘或内存。在另一些实施例中，所述存储器11也可以是所述电子设备1的外部存储设备，例如该电子设备1配备的插接式硬盘，智能存储卡（Smart Media Card, SMC），安全数字（Secure Digital, SD）卡，闪存卡（Flash Card）等。当然，所述存储器11还可以既包括所述电子设备1的内部存储单元也包括其外部存储设备。本实施例中，存储器11通常用于存储安装于所述电子设备1的操作系统和各类应用软件，例如基于人工智能的远程问诊程序10的程序代码等。此外，存储器11还可以用于暂时地存储已经输出或者将要输出的各类数据。
处理器12在一些实施例中可以是中央处理器（Central Processing Unit，CPU）、控制器、微控制器、微处理器、或其他数据处理芯片。该处理器12通常用于控制所述电子设备1的总体操作，例如执行数据交互或者通信相关的控制和处理等。本实施例中，所述处理器12用于运行所述存储器11中存储的程序代码或者处理数据，例如运行基于人工智能的远程问诊程序10的程序代码等。
显示器13可以称为显示屏或显示单元。在一些实施例中显示器13可以是LED显示器、液晶显示器、触控式液晶显示器以及有机发光二极管（Organic Light-Emitting Diode，OLED）触摸器等。显示器13用于显示在电子设备1中处理的信息以及用于显示可视化的工作界面，例如显示数据统计的结果。
网络接口14可选地可以包括标准的有线接口、无线接口（如WI-FI接口），该网络接口14通常用于在所述电子设备1与其它电子设备之间建立通信连接。
图3仅示出了具有组件11-14以及基于人工智能的远程问诊程序10的电子设备1，但是应理解的是，并不要求实施所有示出的组件，可以替代的实施更多或者更少的组件。
可选地，所述电子设备1还可以包括用户接口，用户接口可以包括显示器（Display）、输入单元比如键盘（Keyboard），可选的用户接口还可以包括标准的有线接口、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有机发光二极管（Organic Light-Emitting Diode，OLED）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1中处理的信息以及用于显示可视化的用户界面。
该电子设备1还可以包括射频（Radio Frequency，RF）电路、传感器和音频电路等等，在此不再赘述。
在上述实施例中，处理器12执行存储器11中存储的基于人工智能的远程问诊程序10时可以实现如下步骤：
获取用户利用预设终端上传的问诊信息，基于所述用户的标识判断所述用户所属的用户类型；
当判断所述用户属于基于问诊文本进行问诊的用户时，提取所述问诊信息中问诊文本的关键词，基于所述关键词向所述用户反馈候选科室选项，接收用户基于所述候选科室选项的选项将所述问诊信息反馈至该科室对应的终端；
当判断所述用户属于基于体征数据和图像进行问诊的用户时，将所述问诊信息中用于指示体征数据的指标信息输入预先建立的指标识别模型，得到所述问诊信息的指标识别结果；
提取所述问诊信息中图像的特征信息，将所述特征信息输入预先建立图像识别模型，得到所述问诊信息的图像识别结果，将所述问诊信息、指标识别结果及图像识别结果，发送至所述指标识别结果的类型及所述图像识别结果的类型对应的终端。
所述存储设备可以为电子设备1的存储器11，也可以为与电子设备1通讯连接的其它存储设备。
关于上述步骤的详细介绍，请参照上述图2关于基于人工智能的远程问诊装置100实施例的功能模块图以及图1关于基于人工智能的远程问诊方法实施例的流程图的说明。
此外，本申请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可以是非易失性的，也可以是易失性的。该计算机可读存储介质可以是硬盘、多媒体卡、SD卡、闪存卡、SMC、只读存储器（ROM）、可擦除可编程只读存储器（EPROM）、便携式紧致盘只读存储器（CD-ROM）、USB存储器等等中的任意一种或者几种的任意组合。所述计算机可读存储介质中包括存储数据区和存储程序区，存储数据区存储根据区块链节点的使用所创建的数据，存储程序区存储有基于人工智能的远程问诊程序10，所述基于人工智能的远程问诊程序10被处理器执行时实现如下操作：
获取用户利用预设终端上传的问诊信息，基于所述用户的标识判断所述用户所属的用户类型；
当判断所述用户属于基于问诊文本进行问诊的用户时，提取所述问诊信息中问诊文本的关键词，基于所述关键词向所述用户反馈候选科室选项，接收用户基于所述候选科室选项的选项将所述问诊信息反馈至该科室对应的终端；
当判断所述用户属于基于体征数据和图像进行问诊的用户时，将所述问诊信息中用于指示体征数据的指标信息输入预先建立的指标识别模型，得到所述问诊信息的指标识别结果；
提取所述问诊信息中图像的特征信息，将所述特征信息输入预先建立图像识别模型，得到所述问诊信息的图像识别结果，将所述问诊信息、指标识别结果及图像识别结果，发送至所述指标识别结果的类型及所述图像识别结果的类型对应的终端。
本申请之计算机可读存储介质的具体实施方式与上述基于人工智能的远程问诊方法的具体实施方式大致相同，在此不再赘述。
在另一个实施例中，本申请所提供的基于人工智能的远程问诊方法，为进一步保证上述所有出现的数据的私密和安全性，上述所有数据还可以存储于一区块链的节点中。例如指标识别结果及图像识别结果等，这些数据均可存储在区块链节点中。
需要说明的是，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。
需要说明的是，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。并且本文中的术语“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、装置、物品或者方法不仅包括那些要素，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、装置、物品或者方法所固有的要素。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、装置、物品或者方法中还存在另外的相同要素。
通过以上的实施方式的描述，本领域的技术人员可以清楚地了解到上述实施例方法可借助软件加必需的通用硬件平台的方式来实现，当然也可以通过硬件，但很多情况下前者是更佳的实施方式。基于这样的理解，本申请的技术方案本质上或者说对现有技术做出贡献的部分可以以软件产品的形式体现出来，该计算机软件产品存储在如上所述的一个存储介质（如ROM/RAM、磁碟、光盘）中，包括若干指令用以使得一台终端设备（可以是手机，计算机，电子装置，或者网络设备等）执行本申请各个实施例所述的方法。
以上仅为本申请的优选实施例，并非因此限制本申请的专利范围，凡是利用本申请说明书及附图内容所作的等效结构或等效流程变换，或直接或间接运用在其他相关的技术领域，均同理包括在本申请的专利保护范围内。

Claims (20)

  1. 一种基于人工智能的远程问诊方法，应用于电子设备，其中，所述方法包括：
    获取用户利用预设终端上传的问诊信息，基于所述用户的标识判断所述用户所属的用户类型；
    当判断所述用户属于基于问诊文本进行问诊的用户时，提取所述问诊信息中问诊文本的关键词，基于所述关键词向所述用户反馈候选科室选项，接收用户基于所述候选科室选项的选项将所述问诊信息反馈至该科室对应的终端；
    当判断所述用户属于基于体征数据和图像进行问诊的用户时，将所述问诊信息中用于指示体征数据的指标信息输入预先建立的指标识别模型，得到所述问诊信息的指标识别结果；
    提取所述问诊信息中图像的特征信息，将所述特征信息输入预先建立图像识别模型，得到所述问诊信息的图像识别结果，将所述问诊信息、指标识别结果及图像识别结果，发送至所述指标识别结果的类型及所述图像识别结果的类型对应的终端。
  2. 如权利要求1所述的基于人工智能的远程问诊方法，其中，在基于所述用户的标识判断所述用户所属的用户类型之后，所述方法还包括：
    当判断所述用户属于未注册账号的用户时，将预设的注册界面反馈至所述用户，以供所述用户基于所述注册界面输入并上传所述用户的基本信息进行注册。
  3. 如权利要求1所述的基于人工智能的远程问诊方法，其中，所述提取所述问诊信息中问诊文本的关键词，包括：
    获取所述问诊信息中的问诊文本，并对所述问诊文本执行分词操作得到多个分词；
    计算各分词在所述问诊文本中的词频，基于所述词频计算出各分词的IDF值及TF值，将各分词的IDF值与各分词对应的TF值相乘得到各分词的TF-IDF值；
    基于各分词的TF-IDF值选取第一预设数量的预设词性的词作为所述关键词。
  4. 如权利要求1所述的基于人工智能的远程问诊方法，其中，所述基于所述关键词向所述用户反馈候选科室选项，包括：
    将所述关键词分别与预设的科室语料表对应的标签集进行匹配得到多个标签集匹配度值，选取出标签集匹配度值最高的第二预设数量的候选标签集，将所述问诊文本分别与所述候选标签集对应科室语料表中的各语句进行匹配得到多个语句匹配度值；
    当科室语料表中存在语句匹配度值大于第一预设阈值的语句时，将该语句对应的科室作为所述候选科室，并将语句匹配度值最高的语句对应的候选科室标注预设推荐信息。
  5. 如权利要求1所述的基于人工智能的远程问诊方法，其中，所述指标识别模型的训练过程包括：
    获取第三预设数量的指标数据集，对各指标数据集执行预处理操作后，分别为各指标数据集分配预设标签，将指标数据集作为自变量、指标数据集对应的预设标签作为因变量生成训练样本集；
    将所述训练样本集按照预设比例分成训练集及验证集；
    利用所述训练集中的各自变量及各因变量对卷积神经网络模型进行训练，每隔预设周期利用所述验证集中各变量及各因变量对卷积神经网络模型的准确率进行验证；及
    当验证所述准确率大于第二预设阈值时，结束训练得到所述指标识别模型。
  6. 如权利要求1所述的基于人工智能的远程问诊方法，其中，所述提取所述问诊信息中图像的特征信息，包括：
    利用LBP算法提取出所述图像的纹理特征，将所述图像裁剪成若干张裁剪图像，利用颜色矩提取算法提取出各裁剪图像的颜色特征，对所述纹理特征及各裁剪图像颜色特征执行融合操作得到所述图像的融合特征，将所述融合特征作为所述图像的特征信息。
  7. 如权利要求6所述的基于人工智能的远程问诊方法，其中，所述将所述图像裁剪成若干张裁剪图像，包括：
    在所述图像的左下角设置坐标原点，将所述图像分别沿着x轴和y轴方向，且以预设步长进行裁剪，得到若干张对应的预设大小的裁剪图像。
  8. 一种基于人工智能的远程问诊装置，其中，所述装置包括：
    判断模块：用于获取用户利用预设终端上传的问诊信息，基于所述用户的标识判断所述用户所属的用户类型；
    第一处理模块：用于当判断所述用户属于基于问诊文本进行问诊的用户时，提取所述问诊信息中问诊文本的关键词，基于所述关键词向所述用户反馈候选科室选项，接收用户基于所述候选科室选项的选项将所述问诊信息反馈至该科室对应的终端；
    第二处理模块：用于当判断所述用户属于基于体征数据和图像进行问诊的用户时，将所述问诊信息中用于指示体征数据的指标信息输入预先建立的指标识别模型，得到所述问诊信息的指标识别结果；
    发送模块：用于提取所述问诊信息中图像的特征信息，将所述特征信息输入预先建立图像识别模型，得到所述问诊信息的图像识别结果，将所述问诊信息、指标识别结果及图像识别结果，发送至所述指标识别结果的类型及所述图像识别结果的类型对应的终端。
  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程序，所述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远程问诊方法：
    获取用户利用预设终端上传的问诊信息，基于所述用户的标识判断所述用户所属的用户类型；
    当判断所述用户属于基于问诊文本进行问诊的用户时，提取所述问诊信息中问诊文本的关键词，基于所述关键词向所述用户反馈候选科室选项，接收用户基于所述候选科室选项的选项将所述问诊信息反馈至该科室对应的终端；
    当判断所述用户属于基于体征数据和图像进行问诊的用户时，将所述问诊信息中用于指示体征数据的指标信息输入预先建立的指标识别模型，得到所述问诊信息的指标识别结果；
    提取所述问诊信息中图像的特征信息，将所述特征信息输入预先建立图像识别模型，得到所述问诊信息的图像识别结果，将所述问诊信息、指标识别结果及图像识别结果，发送至所述指标识别结果的类型及所述图像识别结果的类型对应的终端。
  10. 如权利要求9所述的电子设备，其中，在基于所述用户的标识判断所述用户所属的用户类型之后，所述方法还包括：
    当判断所述用户属于未注册账号的用户时，将预设的注册界面反馈至所述用户，以供所述用户基于所述注册界面输入并上传所述用户的基本信息进行注册。
  11. 如权利要求9所述的电子设备，其中，所述提取所述问诊信息中问诊文本的关键词，包括：
    获取所述问诊信息中的问诊文本，并对所述问诊文本执行分词操作得到多个分词；
    计算各分词在所述问诊文本中的词频，基于所述词频计算出各分词的IDF值及TF值，将各分词的IDF值与各分词对应的TF值相乘得到各分词的TF-IDF值；
    基于各分词的TF-IDF值选取第一预设数量的预设词性的词作为所述关键词。
  12. 如权利要求9所述的电子设备，其中，所述基于所述关键词向所述用户反馈候选科室选项，包括：
    将所述关键词分别与预设的科室语料表对应的标签集进行匹配得到多个标签集匹配度值，选取出标签集匹配度值最高的第二预设数量的候选标签集，将所述问诊文本分别与所述候选标签集对应科室语料表中的各语句进行匹配得到多个语句匹配度值；
    当科室语料表中存在语句匹配度值大于第一预设阈值的语句时，将该语句对应的科室作为所述候选科室，并将语句匹配度值最高的语句对应的候选科室标注预设推荐信息。
  13. 如权利要求9所述的电子设备，其中，所述指标识别模型的训练过程包括：
    获取第三预设数量的指标数据集，对各指标数据集执行预处理操作后，分别为各指标数据集分配预设标签，将指标数据集作为自变量、指标数据集对应的预设标签作为因变量生成训练样本集；
    将所述训练样本集按照预设比例分成训练集及验证集；
    利用所述训练集中的各自变量及各因变量对卷积神经网络模型进行训练，每隔预设周期利用所述验证集中各变量及各因变量对卷积神经网络模型的准确率进行验证；及
    当验证所述准确率大于第二预设阈值时，结束训练得到所述指标识别模型。
  14. 如权利要求9所述的电子设备，其中，所述提取所述问诊信息中图像的特征信息，包括：
    利用LBP算法提取出所述图像的纹理特征，将所述图像裁剪成若干张裁剪图像，利用颜色矩提取算法提取出各裁剪图像的颜色特征，对所述纹理特征及各裁剪图像颜色特征执行融合操作得到所述图像的融合特征，将所述融合特征作为所述图像的特征信息。
  15. [根据细则26改正11.07.2022]
    一种计算机可读存储介质，其中，所述计算机可读存储介质存储有基于人工智能的远程问诊程序，所述基于人工智能的远程问诊程序被处理器执行时，实现如下所述基于人工智能的远程问诊方法的步骤：
    获取用户利用预设终端上传的问诊信息，基于所述用户的标识判断所述用户所属的用户类型；
    当判断所述用户属于基于问诊文本进行问诊的用户时，提取所述问诊信息中问诊文本的关键词，基于所述关键词向所述用户反馈候选科室选项，接收用户基于所述候选科室选项的选项将所述问诊信息反馈至该科室对应的终端；
    当判断所述用户属于基于体征数据和图像进行问诊的用户时，将所述问诊信息中用于指示体征数据的指标信息输入预先建立的指标识别模型，得到所述问诊信息的指标识别结果；
    提取所述问诊信息中图像的特征信息，将所述特征信息输入预先建立图像识别模型，得到所述问诊信息的图像识别结果，将所述问诊信息、指标识别结果及图像识别结果，发送至所述指标识别结果的类型及所述图像识别结果的类型对应的终端。
  16. 如权利要求15所述的计算机可读存储介质，其中，在基于所述用户的标识判断所述用户所属的用户类型之后，所述方法还包括：
    当判断所述用户属于未注册账号的用户时，将预设的注册界面反馈至所述用户，以供所述用户基于所述注册界面输入并上传所述用户的基本信息进行注册。
  17. 如权利要求15所述的计算机可读存储介质，其中，所述提取所述问诊信息中问诊文本的关键词，包括：
    获取所述问诊信息中的问诊文本，并对所述问诊文本执行分词操作得到多个分词；
    计算各分词在所述问诊文本中的词频，基于所述词频计算出各分词的IDF值及TF值，将各分词的IDF值与各分词对应的TF值相乘得到各分词的TF-IDF值；
    基于各分词的TF-IDF值选取第一预设数量的预设词性的词作为所述关键词。
  18. 如权利要求15所述的计算机可读存储介质，其中，所述基于所述关键词向所述用户反馈候选科室选项，包括：
    将所述关键词分别与预设的科室语料表对应的标签集进行匹配得到多个标签集匹配度值，选取出标签集匹配度值最高的第二预设数量的候选标签集，将所述问诊文本分别与所述候选标签集对应科室语料表中的各语句进行匹配得到多个语句匹配度值；
    当科室语料表中存在语句匹配度值大于第一预设阈值的语句时，将该语句对应的科室作为所述候选科室，并将语句匹配度值最高的语句对应的候选科室标注预设推荐信息。
  19. 如权利要求15所述的计算机可读存储介质，其中，所述指标识别模型的训练过程包括：
    获取第三预设数量的指标数据集，对各指标数据集执行预处理操作后，分别为各指标数据集分配预设标签，将指标数据集作为自变量、指标数据集对应的预设标签作为因变量生成训练样本集；
    将所述训练样本集按照预设比例分成训练集及验证集；
    利用所述训练集中的各自变量及各因变量对卷积神经网络模型进行训练，每隔预设周期利用所述验证集中各变量及各因变量对卷积神经网络模型的准确率进行验证；及
    当验证所述准确率大于第二预设阈值时，结束训练得到所述指标识别模型。
  20. 如权利要求15所述的计算机可读存储介质，其中，所述提取所述问诊信息中图像的特征信息，包括：
    利用LBP算法提取出所述图像的纹理特征，将所述图像裁剪成若干张裁剪图像，利用颜色矩提取算法提取出各裁剪图像的颜色特征，对所述纹理特征及各裁剪图像颜色特征执行融合操作得到所述图像的融合特征，将所述融合特征作为所述图像的特征信息。
PCT/CN2022/087815 2021-08-30 2022-04-20 基于人工智能的远程问诊方法、装置、设备及介质 WO2023029510A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202111003077.0 2021-08-30
CN202111003077.0A CN113707301A (zh) 2021-08-30 2021-08-30 基于人工智能的远程问诊方法、装置、设备及介质

Publications (1)

Publication Number Publication Date
WO2023029510A1 true WO2023029510A1 (zh) 2023-03-09

Family

ID=78656675

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2022/087815 WO2023029510A1 (zh) 2021-08-30 2022-04-20 基于人工智能的远程问诊方法、装置、设备及介质

Country Status (2)

Country Link
CN (1) CN113707301A (zh)
WO (1) WO2023029510A1 (zh)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116110574A (zh) * 2023-04-14 2023-05-12 武汉大学人民医院（湖北省人民医院） 一种基于神经网络实现的眼科智能问诊方法和装置

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113707301A (zh) * 2021-08-30 2021-11-26 康键信息技术（深圳）有限公司 基于人工智能的远程问诊方法、装置、设备及介质

Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR20140054660A (ko) * 2012-10-29 2014-05-09 한국디지털병원수출사업협동조합 3차원 초음파 진단기를 이용한 원격자문 시스템
CN109166618A (zh) * 2017-06-28 2019-01-08 京东方科技集团股份有限公司 分诊系统和分诊方法
KR101981326B1 (ko) * 2018-12-14 2019-05-22 가천대학교 산학협력단 로봇 인지용 플랫폼, 그 시스템 및 그 진단 방법
WO2019200535A1 (zh) * 2018-04-17 2019-10-24 深圳华大生命科学研究院 基于人工智能的眼科疾病诊断建模方法、装置及系统
CN111241265A (zh) * 2020-01-09 2020-06-05 平安国际智慧城市科技股份有限公司 信息推荐方法、设备、存储介质及装置
CN112168141A (zh) * 2020-01-20 2021-01-05 广州海鹚网络科技有限公司 远程问诊的信息处理方法、装置、设备和存储介质
CN112509682A (zh) * 2020-12-15 2021-03-16 康键信息技术（深圳）有限公司 基于文本识别的问诊方法、装置、设备及存储介质
CN113707301A (zh) * 2021-08-30 2021-11-26 康键信息技术（深圳）有限公司 基于人工智能的远程问诊方法、装置、设备及介质

Family Cites Families (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108986908B (zh) * 2018-05-31 2023-04-18 平安医疗科技有限公司 问诊数据处理方法、装置、计算机设备和存储介质
CN110634578A (zh) * 2019-09-16 2019-12-31 重庆百行智能数据科技研究院有限公司 一种远程智能问诊的方法
CN111816301A (zh) * 2020-07-07 2020-10-23 平安科技（深圳）有限公司 医疗问诊辅助方法、装置、电子设备及介质
CN112349435A (zh) * 2020-12-08 2021-02-09 微医云(杭州)控股有限公司 一种在线问诊方法、装置、服务器、介质和系统
CN112863701A (zh) * 2021-02-01 2021-05-28 北京好欣晴移动医疗科技有限公司 非接触智能问诊系统

Patent Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR20140054660A (ko) * 2012-10-29 2014-05-09 한국디지털병원수출사업협동조합 3차원 초음파 진단기를 이용한 원격자문 시스템
CN109166618A (zh) * 2017-06-28 2019-01-08 京东方科技集团股份有限公司 分诊系统和分诊方法
WO2019200535A1 (zh) * 2018-04-17 2019-10-24 深圳华大生命科学研究院 基于人工智能的眼科疾病诊断建模方法、装置及系统
KR101981326B1 (ko) * 2018-12-14 2019-05-22 가천대학교 산학협력단 로봇 인지용 플랫폼, 그 시스템 및 그 진단 방법
CN111241265A (zh) * 2020-01-09 2020-06-05 平安国际智慧城市科技股份有限公司 信息推荐方法、设备、存储介质及装置
CN112168141A (zh) * 2020-01-20 2021-01-05 广州海鹚网络科技有限公司 远程问诊的信息处理方法、装置、设备和存储介质
CN112509682A (zh) * 2020-12-15 2021-03-16 康键信息技术（深圳）有限公司 基于文本识别的问诊方法、装置、设备及存储介质
CN113707301A (zh) * 2021-08-30 2021-11-26 康键信息技术（深圳）有限公司 基于人工智能的远程问诊方法、装置、设备及介质

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116110574A (zh) * 2023-04-14 2023-05-12 武汉大学人民医院（湖北省人民医院） 一种基于神经网络实现的眼科智能问诊方法和装置
US11955240B1 (en) 2023-04-14 2024-04-09 Renmin Hospital Of Wuhan University (hubei General Hospital) Neural-network-based-implemented ophthalmologic intelligent consultation method and apparatus

Also Published As

Publication number Publication date
CN113707301A (zh) 2021-11-26

Similar Documents

Publication Publication Date Title
CN112417096B (zh) 问答对匹配方法、装置、电子设备及存储介质
WO2019085330A1 (zh) 人物性格分析方法、装置及存储介质
CN109036545B (zh) 医疗信息处理方法、装置、计算机设备和存储介质
WO2023029510A1 (zh) 基于人工智能的远程问诊方法、装置、设备及介质
CN112447300B (zh) 基于图神经网络的医疗查询方法、装置、计算机设备及存储介质
WO2023029513A1 (zh) 基于人工智能的搜索意图识别方法、装置、设备及介质
WO2021151270A1 (zh) 图像结构化数据提取方法、装置、设备及存储介质
CN111985241B (zh) 医学信息查询方法、装置、电子设备及介质
CN113707303A (zh) 基于知识图谱的医疗问题解答方法、装置、设备及介质
CN112509682A (zh) 基于文本识别的问诊方法、装置、设备及存储介质
CN111581623B (zh) 智能数据交互方法、装置、电子设备及存储介质
CN112035611B (zh) 目标用户推荐方法、装置、计算机设备和存储介质
CN113111162A (zh) 科室推荐方法、装置、电子设备及存储介质
CN114860887A (zh) 基于智能联想的疾病内容推送方法、装置、设备及介质
CN110752027B (zh) 电子病历数据推送方法、装置、计算机设备和存储介质
CN113111159A (zh) 问答记录生成方法、装置、电子设备及存储介质
CN113724830B (zh) 基于人工智能的用药风险检测方法及相关设备
CN115346634A (zh) 一种体检报告解读预测方法、系统、电子设备和存储介质
CN116578704A (zh) 文本情感分类方法、装置、设备及计算机可读介质
CN112151187A (zh) 信息查询方法、装置、计算机设备和存储介质
CN111933241B (zh) 医疗数据解析方法、装置、电子设备及存储介质
CN113688119B (zh) 基于人工智能的医疗数据库构建方法及相关设备
CN111931091B (zh) 页面内容展示方法、装置、计算机设备及存储介质
CN115374192A (zh) 一种排序方法、装置、电子设备及存储介质
CN108511044B (zh) 一种互联网咨询分诊方法及系统

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: 22862656

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE