CN113488159A - Medical department recommendation method and device based on neural network - Google Patents
Medical department recommendation method and device based on neural network Download PDFInfo
- Publication number
- CN113488159A CN113488159A CN202110916275.XA CN202110916275A CN113488159A CN 113488159 A CN113488159 A CN 113488159A CN 202110916275 A CN202110916275 A CN 202110916275A CN 113488159 A CN113488159 A CN 113488159A
- Authority
- CN
- China
- Prior art keywords
- disease
- inquiry
- entity
- disease entity
- medical department
- Prior art date
- Legal status (The legal status 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 status listed.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H40/00—ICT specially adapted for the management or administration of healthcare resources or facilities; ICT specially adapted for the management or operation of medical equipment or devices
- G16H40/20—ICT specially adapted for the management or administration of healthcare resources or facilities; ICT specially adapted for the management or operation of medical equipment or devices for the management or administration of healthcare resources or facilities, e.g. managing hospital staff or surgery rooms
-
- 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
- G06F16/335—Filtering based on additional data, e.g. user or group profiles
-
- 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
- G06F16/353—Clustering; Classification into predefined classes
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/044—Recurrent networks, e.g. Hopfield networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/048—Activation functions
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Physics & Mathematics (AREA)
- Biomedical Technology (AREA)
- Data Mining & Analysis (AREA)
- General Health & Medical Sciences (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Computational Linguistics (AREA)
- Computing Systems (AREA)
- Public Health (AREA)
- Evolutionary Computation (AREA)
- Biophysics (AREA)
- Artificial Intelligence (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Life Sciences & Earth Sciences (AREA)
- Databases & Information Systems (AREA)
- Molecular Biology (AREA)
- Medical Informatics (AREA)
- Epidemiology (AREA)
- Primary Health Care (AREA)
- General Business, Economics & Management (AREA)
- Business, Economics & Management (AREA)
- Pathology (AREA)
- Medical Treatment And Welfare Office Work (AREA)
Abstract
The application relates to the field of intelligent decision making, and discloses a medical department recommendation method based on a neural network, which comprises the following steps: receiving inquiry data input by a user, and performing feature extraction on the inquiry data to obtain feature inquiry data; identifying a disease label of the characteristic inquiry data by using a label classification network of a disease entity detection model, and identifying a disease entity of the disease label by using an entity regression network of the disease entity detection model to obtain a first disease entity; searching a second disease entity matched with the characteristic inquiry data from the inquiry database; selecting disease entities with the same disease type from the first disease entity and the second disease entity to obtain a target disease entity; and matching the target disease entity with the medical departments in the medical department library, and recommending the successfully matched medical departments to the user. In addition, the application also provides a medical department recommending device based on the neural network, electronic equipment and a storage medium. The recommendation accuracy of the medical department can be improved.
Description
Technical Field
The present application relates to the field of intelligent decision making, and in particular, to a medical department recommendation method and apparatus based on a neural network, an electronic device, and a computer-readable storage medium.
Background
With the continuous development and perfection of artificial intelligence technology, people's daily life is greatly enriched and facilitated, and in the medical field, machines including intelligent diagnosis guide service are configured in many hospitals at present, aiming at realizing the online intelligent diagnosis guide function by a data-driven method through an internet technology platform and machine learning and statistical learning methods, and recommending departments to patients, so that the subjective motility of the personnel in need of treatment is brought into full play, and the personnel in need of the medical service can be effectively and accurately found.
Currently, department recommendation is usually implemented based on a recommendation mode constructed by rules, that is, inquiry information of a user is guided step by step through an intelligent diagnosis guiding machine or program, so as to output a corresponding department for treatment, but such a recommendation mode is easily influenced by the rules set by the intelligent diagnosis guiding machine or program, that is, the user describes own physical condition according to the steps set by the intelligent diagnosis guiding machine or program, and physical data not related to the intelligent diagnosis guiding machine or program is easily ignored, so that the accuracy of department recommendation is influenced.
Disclosure of Invention
In order to solve the technical problems or at least partially solve the technical problems, the present application provides a medical department recommendation method, device, electronic device and computer-readable storage medium based on a neural network, which can improve the recommendation accuracy of a medical department.
In a first aspect, the present application provides a medical department recommendation method based on a neural network, including:
receiving inquiry data input by a user, and performing feature extraction on the inquiry data to obtain feature inquiry data;
identifying a disease label of the characteristic inquiry data by using a label classification network in a disease entity detection model, and identifying a disease entity of the disease label by using an entity regression network in the disease entity detection model to obtain a first disease entity;
searching a second disease entity matched with the characteristic inquiry data from an inquiry database;
selecting disease entities with the same disease type from the first disease entity and the second disease entity to obtain a target disease entity;
and matching the target disease entity with medical departments in a medical department library, and recommending the successfully matched medical departments to the user.
It can be seen that, in the embodiment of the application, by performing feature extraction on the inquiry data, the key information in the inquiry data can be screened out, so that the processing speed of the subsequent inquiry data is increased; secondly, end-to-end identification of a first disease entity can be realized through a disease entity detection model comprising a label classification network and an entity regression network, and model robustness of disease entity identification is guaranteed, so that accuracy of disease identification can be improved, accuracy of follow-up medical department recommendation can be improved, and a premise of disease entity comparison with the first disease entity can be guaranteed by searching a second disease entity matched with the characteristic inquiry data from an inquiry database; further, in the embodiment of the application, disease entities with the same disease type are selected from the first disease entity and the second disease entity, so that the same disease entity in the first disease entity and the second disease entity is used as a target disease entity, and the accuracy of recommendation of medical departments is further guaranteed.
In a possible implementation manner of the first aspect, the performing feature extraction on the inquiry data to obtain feature inquiry data includes:
performing word segmentation on the inquiry data to obtain an inquiry word set;
calculating an information gain value of each inquiry word in the inquiry data in the inquiry word set, and selecting the inquiry words of which the information gain values are larger than a preset gain value;
and generating characteristic inquiry data according to the selected inquiry words.
In one possible implementation manner of the first aspect, the identifying the disease label of the feature inquiry data by using a label classification network in a disease entity detection model includes:
carrying out position vector coding on the feature inquiry data by utilizing a coding layer in the label classification network model to generate a coding inquiry vector;
extracting a tag sequence of the coded inquiry vector by using an attention mechanism in the tag classification network to obtain a tag sequence inquiry vector;
and detecting the label information of the label sequence inquiry vector by using a full connection layer in the label classification network to obtain a disease label.
In a possible implementation manner of the first aspect, the extracting a tag sequence from the encoded inquiry vector by using an attention mechanism in the tag classification network to obtain a tag sequence inquiry vector includes:
performing convolution on the inquiry coding vector by utilizing a convolution module in the attention mechanism to obtain a convolution inquiry vector;
performing label information coding on the convolution inquiry vector by using a coder in the attention mechanism to obtain a label inquiry vector;
and decoding the tag sequence of the tag inquiry vector by using a decoder in the attention mechanism to obtain the tag sequence inquiry vector.
In one possible implementation manner of the first aspect, the identifying the disease entity of the disease signature using an entity regression network in the disease entity detection model to obtain a first disease entity includes:
calculating the state value of the disease label by using an input gate in the entity regression network, and calculating the activation value of the disease label by using a forgetting gate in the entity regression network;
calculating a state update value of the disease label according to the state value and the activation value;
and calculating the disease entity sequence of the state update value by using an output gate in the entity regression network to obtain a first disease entity of the disease label.
In one possible implementation manner of the first aspect, the searching for the second disease entity matching the characteristic inquiry data from the inquiry database includes:
acquiring an inquiry symptom field of the characteristic inquiry data and a disease symptom field in the inquiry database, and matching the inquiry symptom field with the disease symptom field;
and taking the disease entity corresponding to the successfully matched disease symptom field as the disease entity of the inquiry symptom field to obtain a second disease entity of the characteristic inquiry data.
In one possible implementation manner of the first aspect, the matching the target disease entity with a medical department in a medical department library includes:
calculating the matching degree of the target disease entity and medical departments in a medical department library;
if the department matching degree does not meet the preset condition, the target disease entity fails to be matched with the medical department in the medical department library;
and if the department matching degree meets the preset condition, successfully matching the target disease entity with the medical department in the medical department library.
In a second aspect, the present application provides a neural network-based medical department recommendation apparatus, the apparatus comprising:
the data feature extraction module is used for receiving inquiry data input by a user and extracting features of the inquiry data to obtain feature inquiry data;
a disease entity identification module, configured to identify a disease tag of the feature inquiry data by using a tag classification network in a disease entity detection model, and identify a disease entity of the disease tag by using an entity regression network in the disease entity detection model, so as to obtain a first disease entity;
the disease entity matching module is used for searching a second disease entity matched with the characteristic inquiry data from the inquiry database;
a disease entity selection module for selecting disease entities with the same disease type from the first disease entity and the second disease entity to obtain a target disease entity;
and the medical department recommending module is used for matching the target disease entity with medical departments in a medical department library and recommending the successfully matched medical departments to the user.
In a third aspect, the present application provides an electronic device, comprising:
at least one processor; and a memory communicatively coupled to the at least one processor;
wherein the memory stores a computer program executable by the at least one processor to enable the at least one processor to perform the neural network-based medical department recommendation method of any one of the first aspects above.
In a fourth aspect, the present application provides a computer readable storage medium storing a computer program which, when executed by a processor, implements the neural network-based medical department recommendation method as described in any one of the first aspects above.
It is understood that the beneficial effects of the second to fourth aspects can be seen from the description of the first aspect, and are not described herein again.
Drawings
The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and together with the description, serve to explain the principles of the application.
In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings needed to be used in the description of the embodiments or the prior art will be briefly described below, and it is obvious for those skilled in the art to obtain other drawings without inventive exercise.
Fig. 1 is a detailed flowchart of a medical department recommendation method based on a neural network according to an embodiment of the present application.
Fig. 2 is a flowchart illustrating a step of the neural network-based medical department recommendation method illustrated in fig. 1 according to an embodiment of the present application.
Fig. 3 is a flowchart illustrating another step of the neural network-based medical department recommendation method illustrated in fig. 1 according to an embodiment of the present application.
Fig. 4 is a flowchart illustrating another step of the neural network-based medical department recommendation method illustrated in fig. 1 according to an embodiment of the present application.
Fig. 5 is a block diagram of a neural network-based medical department recommendation device according to an embodiment of the present application.
Fig. 6 is a schematic internal structural diagram of an electronic device for implementing a neural network-based medical department recommendation method according to an embodiment of the present application.
Detailed Description
In order to make the objects, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it is obvious that the described embodiments are some embodiments of the present application, but not all embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present application.
A neural network-based medical department recommendation method provided by an embodiment of the present application is described with reference to a flowchart shown in fig. 1. The neural network-based medical department recommendation method described in fig. 1 includes:
and S1, receiving inquiry data input by a user, and performing feature extraction on the inquiry data to obtain feature inquiry data.
In this embodiment of the application, the inquiry data may be understood as personal information and behavior information input by a user on an intelligent inquiry platform, and the intelligent inquiry platform may be an app program, an intelligent inquiry machine, a public number, and the like, where the personal information refers to basic identity information of the user, and includes: name, gender, contact information, age and the like, wherein the behavior information refers to the physical state change data of the user, and comprises: mental state description data (such as dizziness, heaviness of head and feet, chest distress and the like), physical symptom description data (such as sweating caused by dizziness, nausea and dizziness caused by eating bad things) and the like. It should be understood that there may be many useless data in the inquiry data, and if there is inquiry data including "light-up and dizziness found in the light-up table of today", the "getting-up and finding" in the inquiry data may be defined as useless data, so the embodiments of the present application may improve the processing speed of the subsequent inquiry data by performing feature extraction on the inquiry data to screen out the key information in the inquiry data.
As an embodiment of the present application, referring to fig. 2, the performing feature extraction on the inquiry data to obtain feature inquiry data includes:
s201, performing word segmentation on the inquiry data to obtain an inquiry word set;
s202, calculating an information gain value of each inquiry word in the inquiry data in the inquiry word set, and selecting the inquiry words of which the information gain values are larger than a preset gain value;
and S203, generating characteristic inquiry data according to the selected inquiry words.
In an alternative embodiment, the segmentation of the inquiry data may be implemented by a segmentation algorithm, such as a final segmentation algorithm, a dictionary segmentation algorithm, a markov segmentation algorithm, and the like.
In an alternative embodiment, the information gain value may be understood as an information weighted value of the inquiry words in the inquiry data, and in the present application, the information gain value of each inquiry word in the inquiry data in the inquiry word set is calculated by using the following formula:
wherein the content of the first and second substances,an information gain value representing the ith interrogation word in the set of interrogation words,words for presentation and inquiryThe ith word of inquiry is collected,representing the feature vector covariance of the ith query term in the set of query terms, trace () representing the spatial filter function.
In an optional embodiment, the preset gain value may be set to 0.6, or may be set according to an actual service scenario.
S2, recognizing the disease label of the characteristic inquiry data by using a label classification network in a disease entity detection model, and recognizing the disease entity of the disease label by using an entity regression network in the disease entity detection model to obtain a first disease entity.
In the embodiment of the present application, the disease entity detection model includes a tag classification network and an entity regression network, and is used for identifying disease entities (such as gallstones, lung nodules, cold fever, etc.) in the characteristic inquiry data, wherein the tag classification network is constructed by a Convolutional Neural Network (CNN) and is used for detecting disease tags (such as dizziness, nausea, chest distress, etc.) in the characteristic inquiry data, the entity regression network is constructed by a Long Short-Term Memory network (LSTM) and is used for identifying disease names of disease tags and determining the disease entities in the characteristic inquiry data, and the disease entity detection model constructed based on the tag classification network and the entity regression network can realize end-to-end detection of the disease entities in the characteristic inquiry data, and ensure the robustness of the model for identifying the disease entities, therefore, the accuracy of disease identification can be improved, and the accuracy of follow-up medical department recommendation can be further improved.
Referring to fig. 3, as an embodiment of the present application, the identifying the disease label of the feature inquiry data by using a label classification network in a disease entity detection model includes:
s301, carrying out position vector coding on the feature inquiry data by utilizing a coding layer in the label classification network model to generate a coding inquiry vector;
s302, extracting a tag sequence of the coded inquiry vector by using an attention mechanism in the tag classification network to obtain a tag sequence inquiry vector;
s303, detecting the label information of the label sequence inquiry vector by using a full connection layer in the label classification network to obtain a disease label.
In an alternative embodiment, the encoding layer 301 comprises: index coding is carried out on characters in the characteristic inquiry data by utilizing the coding layer to obtain character coding indexes; converting the characters into corresponding character vectors by utilizing the coding layer to obtain initial character vectors; and combining the character coding index and the character vector to generate a coding inquiry vector.
In an alternative embodiment, the attention mechanism comprises: a convolution module, an encoder, and a decoder, the S302 includes: performing convolution on the inquiry coding vector by utilizing a convolution module in the attention mechanism to obtain a convolution inquiry vector; performing label information coding on the convolution inquiry vector by using a coder in the attention mechanism to obtain a label inquiry vector; and decoding the tag sequence of the tag inquiry vector by using a decoder in the attention mechanism to obtain the tag sequence inquiry vector.
In an alternative embodiment, the fully-connected layer comprises: activating a function and feedforward neural network, the S303 including: and detecting the disease label information of the label sequence inquiry vector by using the activation function in the full connection layer, and outputting the disease label information by using the feedforward neural network of the full connection layer to obtain a disease label.
Based on the detection of the disease label, the disease types in the characteristic inquiry data can be identified, and the disease classification of the inquiry data is realized, so that the identification premise of subsequent disease entities is guaranteed.
Further, referring to fig. 4 as an embodiment of the present application, the identifying the disease entity of the disease label by using the entity regression network in the disease entity detection model to obtain a first disease entity includes:
s401, calculating a state value of the disease label by using an input gate in the entity regression network, and calculating an activation value of the disease label by using a forgetting gate in the entity regression network;
s402, calculating a state update value of the disease label according to the state value and the activation value;
s403, calculating a disease entity sequence of the state update value by using an output gate in the entity regression network to obtain a first disease entity of the disease label.
In an alternative embodiment, the status value of the disease tag is calculated using the following formula:
wherein the content of the first and second substances,the value of the state is represented by,indicating the offset of the cell units in the input gates,indicating the activation factor of the input gate,
represents the peak value of the disease tag at time t-1 of the input gate,indicating that the target sensitive text is at time t,representing the weight of the cell units in the input gate.
In an alternative embodiment, the activation value of the disease tag is calculated using the following formula:
wherein the content of the first and second substances,the value of the activation is represented by,indicating the bias of the cell unit in the forgetting gate,an activation factor that indicates that the door was forgotten,represents the peak value of the disease label at the moment t-1 of the forgetting gate,indicates the disease label entered at time t,representing the weight of the cell unit in the forgetting gate.
In an alternative embodiment, the state update value of the disease tag is calculated using the following formula:
wherein the content of the first and second substances,the value of the state update is represented,represents the peak value of the disease tag at time t-1 of the input gate,indicating the peak of the disease label at time t-1 of forgetting gate.
In an alternative embodiment, the sequence of disease entities for the state update value is calculated using the following formula:
wherein the content of the first and second substances,represents the sequence of the disease entity or entities,the activation function of the output gate is represented,representing the state update value.
Based on the identification of the first disease entity, the disease information of the characteristic inquiry data can be identified, and the disease division of the inquiry data is realized, so that the follow-up medical department recommendation premise is guaranteed.
It should be noted that, in the embodiment of the present application, the disease entity detection model refers to a model that has been trained in advance, and has a strong disease recognition capability.
And S3, searching a second disease entity matched with the characteristic inquiry data from the inquiry database.
In the embodiment of the present application, the inquiry database refers to a database in which disease entities are established by medical experts according to disease symptoms of users in an actual business scenario, and it can be understood that a relationship between a disease symptom and a disease entity exists in the inquiry database, that is, the disease entity corresponding to the disease symptom can be queried through the corresponding disease symptom. Therefore, according to the embodiment of the application, the second disease entity matched with the characteristic inquiry data is searched from the inquiry database, so that the recommendation accuracy of a subsequent medical department can be guaranteed before the disease entity comparison with the first disease entity is carried out.
As an embodiment of the present application, the searching for the second disease entity matching the characteristic inquiry data from the inquiry database includes: and acquiring an inquiry symptom field of the characteristic inquiry data and a disease symptom field in the inquiry database, matching the inquiry symptom field with the disease symptom field, and taking a disease entity corresponding to the successfully matched disease symptom field as the disease entity of the inquiry symptom field to obtain a second disease entity of the characteristic inquiry data.
Wherein, the inquiry symptom field refers to the information attribute of symptom description data existing in the characteristic inquiry data, and the disease symptom field refers to the information attribute of symptom description data in the inquiry database.
Further, in an optional embodiment of the present application, the matching the inquiry symptom field and the disease symptom field includes: and calculating the similarity between the inquiry symptom field and the disease symptom field, wherein if the similarity is greater than the preset similarity, the inquiry symptom field and the disease symptom field are successfully matched, and if the similarity is not greater than the preset similarity, the inquiry symptom field and the disease symptom field are unsuccessfully matched. The matching degree can be calculated by a cosine similarity algorithm, and the preset similarity can be set to 1 or set according to an actual service scene.
S4, selecting disease entities with the same disease type from the first disease entity and the second disease entity to obtain target disease entities.
It should be understood that the same disease entity and different disease entities may exist in the first disease entity and the second disease entity, and in order to ensure the accuracy of subsequent medical department recommendation, in the embodiments of the present application, the disease entities having the same disease type are selected from the first disease entity and the second disease entity, that is, the same disease entity in the first disease entity and the second disease entity is taken as a target disease entity, so as to further ensure the accuracy of subsequent medical department recommendation.
S5, matching the target disease entity with medical departments in a medical department library, and recommending the successfully matched medical departments to the user.
In an embodiment of the present application, the medical department library is constructed by collecting a plurality of hospital departments, and the medical department includes: surgery (e.g., neurosurgery, dermatology, ophthalmology, etc.), medicine (gastroenterology, respiratory medicine, cardiology, etc.).
As an embodiment of the present application, the matching the target disease entity with a medical department in a medical department library includes: calculating the matching degree of the target disease entity and medical departments in a medical department library; if the department matching degree does not meet the preset condition, the target disease entity fails to be matched with the medical department in the medical department library; and if the department matching degree meets the preset condition, successfully matching the target disease entity with the medical department in the medical department library.
In an alternative implementation, the department matching degree of the target disease entity with the medical department in the medical department library is calculated by using the following formula:
wherein the content of the first and second substances,the matching degree of the department is shown,represents the mth disease entity among the disease entities of interest,represents the nth medical department of the medical department bank.
In an optional implementation, the preset condition may be set as whether the department matching degree is 1, that is, if the department matching degree is 1, the target disease entity is successfully matched with the medical department in the medical department library, and if the department matching degree is not 1, the target disease entity is unsuccessfully matched with the medical department in the medical department library.
According to the method and the device, firstly, by extracting the characteristics of the inquiry data, the key information in the inquiry data can be screened out, and the processing speed of the subsequent inquiry data is improved; secondly, end-to-end identification of a first disease entity can be realized through a disease entity detection model comprising a label classification network and an entity regression network, and model robustness of disease entity identification is guaranteed, so that accuracy of disease identification can be improved, accuracy of follow-up medical department recommendation can be improved, and a premise of disease entity comparison with the first disease entity can be guaranteed by searching a second disease entity matched with the characteristic inquiry data from an inquiry database; further, in the embodiment of the application, disease entities with the same disease type are selected from the first disease entity and the second disease entity, so that the same disease entity in the first disease entity and the second disease entity is used as a target disease entity, and the accuracy of recommendation of medical departments is further guaranteed.
Fig. 5 is a functional block diagram of the neural network-based medical department recommendation device according to the present invention.
The neural network-based medical department recommendation device 500 may be installed in an electronic device. According to the realized function, the medical department recommending device based on the neural network can comprise a data feature extracting module 501, a disease entity identifying module 502, a disease entity matching module 503, a disease entity selecting module 504 and medical department recommending 505 based on the neural network. A module according to the present invention, which may also be referred to as a unit, refers to a series of computer program segments that can be executed by a processor of an electronic device and that can perform a fixed function, and that are stored in a memory of the electronic device.
In the present embodiment, the functions regarding the respective modules/units are as follows:
the data feature extraction module 501 is configured to receive inquiry data input by a user, perform feature extraction on the inquiry data, and obtain feature inquiry data;
the disease entity identification module 502 is configured to identify a disease tag of the feature inquiry data by using a tag classification network in a disease entity detection model, and identify a disease entity of the disease tag by using an entity regression network in the disease entity detection model to obtain a first disease entity;
the disease entity matching module 503 is configured to search a second disease entity matching the characteristic inquiry data from the inquiry database;
the disease entity selecting module 504 is configured to select a disease entity with the same disease type from the first disease entity and the second disease entity to obtain a target disease entity;
the medical department recommending module 505 is configured to match the target disease entity with a medical department in a medical department library, and recommend the successfully matched medical department to the user.
In detail, when the modules in the neural network-based medical department recommendation device 100 in the embodiment of the present application are used, the same technical means as the neural network-based medical department recommendation method described in fig. 1 and 4 above are adopted, and the same technical effects can be produced, and details are not described here again.
Fig. 6 is a schematic structural diagram of an electronic device for implementing the neural network-based medical department recommendation method according to the present application.
The electronic device may include a processor 60, a memory 61, a communication bus 62, and a communication interface 63, and may further include a computer program, such as a neural network-based medical department recommendation program, stored in the memory 61 and executable on the processor 60.
In some embodiments, the processor 60 may be composed of an integrated circuit, for example, a single packaged integrated circuit, or may be composed of a plurality of integrated circuits packaged with the same or different functions, including one or more Central Processing Units (CPUs), microprocessors, digital Processing chips, graphics processors, and combinations of various control chips. The processor 60 is a Control Unit (Control Unit) of the electronic device, connects various components of the whole electronic device by using various interfaces and lines, and executes various functions and processes data of the electronic device by operating or executing programs or modules stored in the memory 61 (for example, executing a neural network-based medical department recommendation program, etc.), and calling data stored in the memory 61.
The memory 61 includes at least one type of readable storage medium including flash memory, removable hard disks, multimedia cards, card-type memory (e.g., SD or DX memory, etc.), magnetic memory, magnetic disks, optical disks, etc. The memory 61 may in some embodiments be an internal storage unit of the electronic device, for example a removable hard disk of the electronic device. The memory 61 may also be an external storage device of the electronic device in other embodiments, such as a plug-in mobile hard disk, a Smart Media Card (SMC), a Secure Digital (SD) Card, a Flash memory Card (Flash Card), and the like, which are provided on the electronic device. Further, the memory 61 may also include both an internal storage unit and an external storage device of the electronic device. The memory 61 may be used not only to store application software installed in the electronic device and various types of data, such as codes of a neural network-based medical department recommended program, etc., but also to temporarily store data that has been output or will be output.
The communication bus 62 may be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus. The bus may be divided into an address bus, a data bus, a control bus, etc. The bus is arranged to enable connection communication between the memory 61 and at least one processor 60 or the like.
The communication interface 63 is used for communication between the electronic device and other devices, and includes a network interface and a user interface. Optionally, the network interface may include a wired interface and/or a wireless interface (e.g., WI-FI interface, bluetooth interface, etc.), which are typically used to establish a communication connection between the electronic device and other electronic devices. The user interface may be a Display (Display), an input unit such as a Keyboard (Keyboard), and optionally a standard wired interface, a wireless interface. Alternatively, in some embodiments, the display may be an LED display, a liquid crystal display, a touch-sensitive liquid crystal display, an OLED (Organic Light-Emitting Diode) touch device, or the like. The display, which may also be referred to as a display screen or display unit, is suitable, among other things, for displaying information processed in the electronic device and for displaying a visualized user interface.
Fig. 6 shows only an electronic device having components, and those skilled in the art will appreciate that the structure shown in fig. 6 does not constitute a limitation of the electronic device, and may include fewer or more components than those shown, or some components may be combined, or a different arrangement of components.
For example, although not shown, the electronic device may further include a power supply (such as a battery) for supplying power to each component, and preferably, the power supply may be logically connected to the at least one processor 60 through a power management device, so that functions of charge management, discharge management, power consumption management and the like are realized through the power management device. The power supply may also include any component of one or more dc or ac power sources, recharging devices, power failure detection circuitry, power converters or inverters, power status indicators, and the like. The electronic device may further include various sensors, a bluetooth module, a Wi-Fi module, and the like, which are not described herein again.
It is to be understood that the described embodiments are for purposes of illustration only and that the scope of the appended claims is not limited to such structures.
The neural network based medical department recommendation program stored by the memory 61 in the electronic device is a combination of a plurality of computer programs that, when executed in the processor 60, may implement:
receiving inquiry data input by a user, and performing feature extraction on the inquiry data to obtain feature inquiry data;
identifying a disease label of the characteristic inquiry data by using a label classification network in a disease entity detection model, and identifying a disease entity of the disease label by using an entity regression network in the disease entity detection model to obtain a first disease entity;
searching a second disease entity matched with the characteristic inquiry data from an inquiry database;
selecting disease entities with the same disease type from the first disease entity and the second disease entity to obtain a target disease entity;
and matching the target disease entity with medical departments in a medical department library, and recommending the successfully matched medical departments to the user.
Specifically, the processor 60 may refer to the description of the relevant steps in the embodiment corresponding to fig. 1 for a specific implementation method of the computer program, which is not described herein again.
Further, the electronic device integrated module/unit, if implemented in the form of a software functional unit and sold or used as a separate product, may be stored in a non-volatile computer-readable storage medium. The computer readable storage medium may be volatile or non-volatile. For example, the computer-readable medium may include: any entity or device capable of carrying said computer program code, recording medium, U-disk, removable hard disk, magnetic disk, optical disk, computer Memory, Read-Only Memory (ROM).
The present application also provides a computer-readable storage medium, storing a computer program that, when executed by a processor of an electronic device, may implement:
receiving inquiry data input by a user, and performing feature extraction on the inquiry data to obtain feature inquiry data;
identifying a disease label of the characteristic inquiry data by using a label classification network in a disease entity detection model, and identifying a disease entity of the disease label by using an entity regression network in the disease entity detection model to obtain a first disease entity;
searching a second disease entity matched with the characteristic inquiry data from an inquiry database;
selecting disease entities with the same disease type from the first disease entity and the second disease entity to obtain a target disease entity;
and matching the target disease entity with medical departments in a medical department library, and recommending the successfully matched medical departments to the user.
In the several embodiments provided in the present application, it should be understood that the disclosed apparatus, device and method may be implemented in other manners. For example, the above-described apparatus embodiments are merely illustrative, and for example, the division of the modules is only one logical functional division, and other divisions may be realized in practice.
The modules described as separate parts may or may not be physically separate, and parts displayed as modules may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of the present embodiment.
In addition, functional modules in the embodiments of the present application may be integrated into one processing unit, or each unit may exist alone physically, or two or more units are integrated into one unit. The integrated unit can be realized in a form of hardware, or in a form of hardware plus a software functional module.
It will be evident to those skilled in the art that the present application is not limited to the details of the foregoing illustrative embodiments, and that the present application may be embodied in other specific forms without departing from the spirit or essential attributes thereof.
The present embodiments are therefore to be considered in all respects as illustrative and not restrictive, the scope of the application being indicated by the appended claims rather than by the foregoing description, and all changes which come within the meaning and range of equivalency of the claims are therefore intended to be embraced therein. Any reference signs in the claims shall not be construed as limiting the claim concerned.
It is noted that, in this document, relational terms such as "first" and "second," and the like, may be used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Also, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other identical elements in a process, method, article, or apparatus that comprises the element.
The above description is merely exemplary of the present application and is presented to enable those skilled in the art to understand and practice the present application. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other embodiments without departing from the spirit or scope of the application. Thus, the present application is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.
Claims (10)
1. A neural network-based medical department recommendation method, the method comprising:
receiving inquiry data input by a user, and performing feature extraction on the inquiry data to obtain feature inquiry data;
identifying a disease label of the characteristic inquiry data by using a label classification network in a disease entity detection model, and identifying a disease entity of the disease label by using an entity regression network in the disease entity detection model to obtain a first disease entity;
searching a second disease entity matched with the characteristic inquiry data from an inquiry database;
selecting disease entities with the same disease type from the first disease entity and the second disease entity to obtain a target disease entity;
and matching the target disease entity with medical departments in a medical department library, and recommending the successfully matched medical departments to the user.
2. The neural network-based medical department recommendation method of claim 1, wherein the performing feature extraction on the interrogation data to obtain feature interrogation data comprises:
performing word segmentation on the inquiry data to obtain an inquiry word set;
calculating an information gain value of each inquiry word in the inquiry data in the inquiry word set, and selecting the inquiry words of which the information gain values are larger than a preset gain value;
and generating characteristic inquiry data according to the selected inquiry words.
3. The neural network-based medical department recommendation method of claim 1, wherein identifying disease tags for the characteristic interrogation data using a tag classification network in a disease entity detection model comprises:
carrying out position vector coding on the feature inquiry data by utilizing a coding layer in the label classification network model to generate a coding inquiry vector;
extracting a tag sequence of the coded inquiry vector by using an attention mechanism in the tag classification network to obtain a tag sequence inquiry vector;
and detecting the label information of the label sequence inquiry vector by using a full connection layer in the label classification network to obtain a disease label.
4. The neural network-based medical department recommendation method of claim 3, wherein the extracting tag sequences from the encoded interrogation vectors using the attention mechanism in the tag classification network to obtain tag sequence interrogation vectors comprises:
performing convolution on the inquiry coding vector by utilizing a convolution module in the attention mechanism to obtain a convolution inquiry vector;
performing label information coding on the convolution inquiry vector by using a coder in the attention mechanism to obtain a label inquiry vector;
and decoding the tag sequence of the tag inquiry vector by using a decoder in the attention mechanism to obtain the tag sequence inquiry vector.
5. The neural network-based medical department recommendation method of claim 1, wherein identifying the disease entity of the disease signature using an entity regression network in the disease entity detection model, resulting in a first disease entity, comprises:
calculating the state value of the disease label by using an input gate in the entity regression network, and calculating the activation value of the disease label by using a forgetting gate in the entity regression network;
calculating a state update value of the disease label according to the state value and the activation value;
and calculating the disease entity sequence of the state update value by using an output gate in the entity regression network to obtain a first disease entity of the disease label.
6. The neural network-based medical department recommendation method of claim 1, wherein said searching for a second disease entity from an interrogation database that matches said characteristic interrogation data comprises:
acquiring an inquiry symptom field of the characteristic inquiry data and a disease symptom field in the inquiry database, and matching the inquiry symptom field with the disease symptom field;
and taking the disease entity corresponding to the successfully matched disease symptom field as the disease entity of the inquiry symptom field to obtain a second disease entity of the characteristic inquiry data.
7. The neural network-based medical department recommendation method of any one of claims 1-6, wherein the matching the target disease entity to a medical department in a medical department repository comprises:
calculating the matching degree of the target disease entity and medical departments in a medical department library;
if the department matching degree does not meet the preset condition, the target disease entity fails to be matched with the medical department in the medical department library;
and if the department matching degree meets the preset condition, successfully matching the target disease entity with the medical department in the medical department library.
8. A neural network-based medical department recommendation device, the device comprising:
the data feature extraction module is used for receiving inquiry data input by a user and extracting features of the inquiry data to obtain feature inquiry data;
a disease entity identification module, configured to identify a disease tag of the feature inquiry data by using a tag classification network in a disease entity detection model, and identify a disease entity of the disease tag by using an entity regression network in the disease entity detection model, so as to obtain a first disease entity;
the disease entity matching module is used for searching a second disease entity matched with the characteristic inquiry data from the inquiry database;
a disease entity selection module for selecting disease entities with the same disease type from the first disease entity and the second disease entity to obtain a target disease entity;
and the medical department recommending module is used for matching the target disease entity with medical departments in a medical department library and recommending the successfully matched medical departments to the user.
9. An electronic device, characterized in that the electronic device comprises:
at least one processor; and the number of the first and second groups,
a memory communicatively coupled to the at least one processor; wherein the content of the first and second substances,
the memory stores a computer program executable by the at least one processor to enable the at least one processor to perform the neural network-based medical department recommendation method of any one of claims 1-7.
10. A computer-readable storage medium storing a computer program, wherein the computer program, when executed by a processor, implements the neural network-based medical department recommendation method of any one of claims 1 to 7.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110916275.XA CN113488159A (en) | 2021-08-11 | 2021-08-11 | Medical department recommendation method and device based on neural network |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110916275.XA CN113488159A (en) | 2021-08-11 | 2021-08-11 | Medical department recommendation method and device based on neural network |
Publications (1)
Publication Number | Publication Date |
---|---|
CN113488159A true CN113488159A (en) | 2021-10-08 |
Family
ID=77946325
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202110916275.XA Pending CN113488159A (en) | 2021-08-11 | 2021-08-11 | Medical department recommendation method and device based on neural network |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN113488159A (en) |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113827274A (en) * | 2021-11-23 | 2021-12-24 | 广东恒腾科技有限公司 | Automatic control system of wireless color Doppler ultrasound equipment |
CN115376668A (en) * | 2022-08-30 | 2022-11-22 | 温州城市智慧健康有限公司 | Big data business analysis method and system applied to intelligent medical treatment |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111028913A (en) * | 2019-11-29 | 2020-04-17 | 北京工业大学 | Hemodialysis treatment scheme aid decision-making method |
CN112037179A (en) * | 2020-08-11 | 2020-12-04 | 深圳大学 | Method, system and equipment for generating brain disease diagnosis model |
CN112885423A (en) * | 2021-03-16 | 2021-06-01 | 康键信息技术(深圳)有限公司 | Disease label detection method and device, electronic equipment and storage medium |
CN113111162A (en) * | 2021-04-21 | 2021-07-13 | 康键信息技术(深圳)有限公司 | Department recommendation method and device, electronic equipment and storage medium |
WO2021139146A1 (en) * | 2020-01-09 | 2021-07-15 | 平安国际智慧城市科技股份有限公司 | Information recommendation method, device, computer-readable storage medium, and apparatus |
-
2021
- 2021-08-11 CN CN202110916275.XA patent/CN113488159A/en active Pending
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111028913A (en) * | 2019-11-29 | 2020-04-17 | 北京工业大学 | Hemodialysis treatment scheme aid decision-making method |
WO2021139146A1 (en) * | 2020-01-09 | 2021-07-15 | 平安国际智慧城市科技股份有限公司 | Information recommendation method, device, computer-readable storage medium, and apparatus |
CN112037179A (en) * | 2020-08-11 | 2020-12-04 | 深圳大学 | Method, system and equipment for generating brain disease diagnosis model |
CN112885423A (en) * | 2021-03-16 | 2021-06-01 | 康键信息技术(深圳)有限公司 | Disease label detection method and device, electronic equipment and storage medium |
CN113111162A (en) * | 2021-04-21 | 2021-07-13 | 康键信息技术(深圳)有限公司 | Department recommendation method and device, electronic equipment and storage medium |
Non-Patent Citations (2)
Title |
---|
李丽双 等: "基于CNN-BLSTM-CRF模型的生物医学命名实体识别", 中文信息学报, vol. 32, no. 1, pages 1 * |
王若佳;魏思仪;王继民;: "BiLSTM-CRF模型在中文电子病历命名实体识别中的应用研究", 文献与数据学报, no. 02 * |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113827274A (en) * | 2021-11-23 | 2021-12-24 | 广东恒腾科技有限公司 | Automatic control system of wireless color Doppler ultrasound equipment |
CN115376668A (en) * | 2022-08-30 | 2022-11-22 | 温州城市智慧健康有限公司 | Big data business analysis method and system applied to intelligent medical treatment |
CN115376668B (en) * | 2022-08-30 | 2024-03-08 | 温州城市智慧健康有限公司 | Big data business analysis method and system applied to intelligent medical treatment |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
EP3985559A1 (en) | Entity semantics relationship classification | |
CN111639498A (en) | Knowledge extraction method and device, electronic equipment and storage medium | |
CN112860848B (en) | Information retrieval method, device, equipment and medium | |
CN113707303A (en) | Method, device, equipment and medium for solving medical problems based on knowledge graph | |
CN113488159A (en) | Medical department recommendation method and device based on neural network | |
CN112231485B (en) | Text recommendation method and device, computer equipment and storage medium | |
CN114648392B (en) | Product recommendation method and device based on user portrait, electronic equipment and medium | |
CN113435202A (en) | Product recommendation method and device based on user portrait, electronic equipment and medium | |
CN111695354A (en) | Text question-answering method and device based on named entity and readable storage medium | |
CN113378970B (en) | Sentence similarity detection method and device, electronic equipment and storage medium | |
CN113707300A (en) | Search intention identification method, device, equipment and medium based on artificial intelligence | |
CN112287069A (en) | Information retrieval method and device based on voice semantics and computer equipment | |
CN113111162A (en) | Department recommendation method and device, electronic equipment and storage medium | |
CN112614578A (en) | Doctor intelligent recommendation method and device, electronic equipment and storage medium | |
CN112132238A (en) | Method, device, equipment and readable medium for identifying private data | |
CN112149409A (en) | Medical word cloud generation method and device, computer equipment and storage medium | |
CN115983271A (en) | Named entity recognition method and named entity recognition model training method | |
CN116030944A (en) | Medical consultation information recommendation method and system | |
CN112860851B (en) | Course recommendation method, device, equipment and medium based on root cause analysis | |
CN112151187B (en) | Information query method, device, computer equipment and storage medium | |
CN113254814A (en) | Network course video labeling method and device, electronic equipment and medium | |
CN113555105A (en) | Method and device for recommending medical products | |
CN114943289A (en) | User portrait classification method, device, equipment and medium based on deep learning | |
CN114596958B (en) | Pathological data classification method, device, equipment and medium based on cascade classification | |
CN113590845B (en) | Knowledge graph-based document retrieval method and device, electronic equipment and medium |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination |