CN117909487A - Medical question-answering service method, system, device and medium for old people - Google Patents
Medical question-answering service method, system, device and medium for old people Download PDFInfo
- Publication number
- CN117909487A CN117909487A CN202410317346.8A CN202410317346A CN117909487A CN 117909487 A CN117909487 A CN 117909487A CN 202410317346 A CN202410317346 A CN 202410317346A CN 117909487 A CN117909487 A CN 117909487A
- Authority
- CN
- China
- Prior art keywords
- medical
- elderly
- knowledge graph
- question
- data
- 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
- 238000000034 method Methods 0.000 title claims abstract description 37
- 238000012549 training Methods 0.000 claims abstract description 16
- 238000011160 research Methods 0.000 claims description 17
- 238000010276 construction Methods 0.000 claims description 13
- 238000002372 labelling Methods 0.000 claims description 11
- 230000007704 transition Effects 0.000 claims description 10
- 230000007246 mechanism Effects 0.000 claims description 9
- 238000005516 engineering process Methods 0.000 claims description 8
- 238000000605 extraction Methods 0.000 claims description 8
- 238000005070 sampling Methods 0.000 claims description 5
- 230000011218 segmentation Effects 0.000 claims description 5
- 238000012795 verification Methods 0.000 claims description 5
- 230000008859 change Effects 0.000 claims description 4
- 238000007781 pre-processing Methods 0.000 claims description 4
- 238000004140 cleaning Methods 0.000 claims description 3
- 230000004044 response Effects 0.000 claims description 3
- 238000013507 mapping Methods 0.000 claims description 2
- 238000013135 deep learning Methods 0.000 abstract description 6
- 238000004891 communication Methods 0.000 description 8
- 230000008569 process Effects 0.000 description 6
- 238000012545 processing Methods 0.000 description 5
- 230000009471 action Effects 0.000 description 3
- 230000008878 coupling Effects 0.000 description 3
- 238000010168 coupling process Methods 0.000 description 3
- 238000005859 coupling reaction Methods 0.000 description 3
- 230000036541 health Effects 0.000 description 3
- 230000006978 adaptation Effects 0.000 description 2
- 239000003814 drug Substances 0.000 description 2
- 230000010354 integration Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000005457 optimization Methods 0.000 description 2
- 238000013473 artificial intelligence Methods 0.000 description 1
- 238000013528 artificial neural network Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008901 benefit Effects 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 238000003745 diagnosis Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 201000010099 disease Diseases 0.000 description 1
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 description 1
- 229940079593 drug Drugs 0.000 description 1
- 238000003058 natural language processing Methods 0.000 description 1
- 230000000474 nursing effect Effects 0.000 description 1
- 230000001575 pathological effect Effects 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 238000012163 sequencing technique Methods 0.000 description 1
- 208000024891 symptom Diseases 0.000 description 1
Classifications
-
- 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
- G06F16/332—Query formulation
- G06F16/3329—Natural language query formulation or dialogue 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
- G06F16/36—Creation of semantic tools, e.g. ontology or thesauri
- G06F16/367—Ontology
-
- 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
- G06F40/295—Named entity recognition
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/30—Semantic analysis
- G06F40/35—Discourse or dialogue representation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/44—Arrangements for executing specific programs
- G06F9/448—Execution paradigms, e.g. implementations of programming paradigms
- G06F9/4498—Finite state machines
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N5/00—Computing arrangements using knowledge-based models
- G06N5/02—Knowledge representation; Symbolic representation
- G06N5/022—Knowledge engineering; Knowledge acquisition
- G06N5/025—Extracting rules from data
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N5/00—Computing arrangements using knowledge-based models
- G06N5/04—Inference or reasoning models
-
- 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
- G16H50/70—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for mining of medical data, e.g. analysing previous cases of other patients
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Health & Medical Sciences (AREA)
- Computational Linguistics (AREA)
- Artificial Intelligence (AREA)
- Software Systems (AREA)
- Mathematical Physics (AREA)
- General Health & Medical Sciences (AREA)
- Medical Informatics (AREA)
- Public Health (AREA)
- Databases & Information Systems (AREA)
- Pathology (AREA)
- Primary Health Care (AREA)
- Epidemiology (AREA)
- Evolutionary Computation (AREA)
- Computing Systems (AREA)
- Biomedical Technology (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Life Sciences & Earth Sciences (AREA)
- Animal Behavior & Ethology (AREA)
- Human Computer Interaction (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention provides a medical question-answering service method, a system, a device and a medium for the elderly, and belongs to the technical field of computers. The method comprises the following steps: constructing an elderly medical knowledge graph; constructing a data set aiming at the language characteristics and medical requirements of the old and training a natural voice understanding model; analyzing user input through a natural voice understanding model to determine query intention, extracting related entities and attributes, and corresponding the extracted entities to the old-fashioned medical knowledge graph; maintaining a conversation history by a finite state machine; constructing a query sentence template aiming at the elderly medical knowledge graph according to user input and intention analysis results; and identifying a corresponding query statement template according to the input information, and querying and outputting an answer in the elderly medical knowledge graph by utilizing the query statement template. According to the invention, by constructing the special senile medical knowledge graph and combining the natural language understanding based on deep learning, the accuracy of medical question and answer is effectively improved.
Description
Technical Field
The invention relates to the technical field of computers, in particular to a medical question-answering service method, a system, a device and a medium for the elderly.
Background
In the medical health field, the application of intelligent question-answering systems is becoming more and more widespread. The basic principle is that the questions of the patient are converted into machine-recognizable languages through natural language processing technology, and accurate answers or suggestions are given according to a medical knowledge base, rules, algorithms and the like. Along with the continuous development of artificial intelligence technology, a medical intelligent question-answering system is also continuously perfected, and the medical intelligent question-answering system has wide application prospect in the aspects of assisting doctors in diagnosis, providing health consultation for patients and the like.
The existing medical question-answering technical scheme mainly comprises the following three steps:
1) Based on information extraction: and extracting information to obtain keywords, and then sequencing the keywords. 2) Based on the general knowledge graph: by constructing a knowledge graph of the entity and its relationship, the question is answered by reasoning on the graph. 3) Based on deep learning: and calculating the question-answer matching degree based on the deep neural network modeling.
Although the above technical solutions can realize intelligent question and answer of medical information, the above technical solutions do not fully consider the specific language and medical requirements of the elderly and the problems of insufficient understanding of complex pathological states when facing the elderly, resulting in insufficient accuracy, low efficiency and slow response speed of medical question and answer.
Disclosure of Invention
Aiming at the problems, the invention aims to provide a medical question-answering service method, a system, a device and a medium for the elderly, which are used for generating answers by constructing a special elderly medical knowledge graph and combining deep learning-based natural language understanding, dialogue management and graph query, so that the accuracy of medical question-answering is effectively improved.
The invention aims to achieve the aim, and the aim is achieved by the following technical scheme: a medical question-answering service method for elderly people, comprising:
Constructing an elderly medical knowledge graph by integrating a medical database, clinical guidelines and elderly medical research data;
constructing a data set aiming at the language characteristics and medical requirements of the old and training a natural voice understanding model;
Analyzing user input through a natural voice understanding model to determine query intention, extracting related entities and attributes, and corresponding the extracted entities to the old-fashioned medical knowledge graph;
Maintaining a dialogue history through a finite state machine to ensure consistency of continuous questions and answers;
Constructing a query sentence template aiming at the elderly medical knowledge graph according to user input and intention analysis results;
And after the user input is obtained, identifying a corresponding query statement template according to the input information, and querying and outputting an answer in the elderly medical knowledge graph by utilizing the query statement template.
Further, by integrating the medical database, the clinical guideline, and the geriatric medical research data, a geriatric medical knowledge graph is constructed, including:
collecting medical data from medical literature, medical encyclopedia, clinical guidelines and geriatric medical research data, and extracting and summarizing medical vocabulary therein using jieba word segmentation tools;
Randomly sampling the extracted vocabulary content, taking 10% of samples for labeling entity types, and determining the relationship among entities;
determining and integrating medical data from medical literature, medical encyclopedia, clinical guidelines and geriatric medical research data, and performing data cleaning and preprocessing to ensure the unification of data formats;
using UIE to extract entities and relations between the entities and the entities;
And verifying the relationship between the entities, and generating an elderly medical knowledge graph after the verification is passed.
Further, constructing a data set for language characteristics and medical requirements of the elderly, and training a natural speech understanding model, comprising:
collecting and organizing a data set containing medical consultation data of the elderly;
analyzing a data set, and determining a problem intention template of the elderly during inquiry according to a preset matching mechanism, wherein the problem intention template comprises entity types and user intentions;
Collecting medical inquiry data of the old, and labeling entities and inquiry intentions in the medical inquiry data;
and training a BERT-based intent recognition and entity recognition joint model by using the labeling result.
Further, analyzing the user input through the natural speech understanding model to determine query intent and extract related entities and attributes, and mapping the extracted entities into a geriatric knowledge graph, including:
Analyzing senile medical inquiry data input by a user through a BERT-based intention recognition and entity recognition combined model, determining inquiry intention and extracting related entities and attributes;
and linking the extracted entity with the corresponding entity in the elderly medical knowledge graph by using a medical term standardization technology.
Further, maintaining a dialogue history by the finite state machine to ensure consistency of continuous questions and answers, including:
Defining states and transition conditions of the states at different stages of the dialogue;
Adding an exception handling mechanism in the finite state machine to handle exceptions or errors occurring during the conversation;
State tracking is performed and state change and transition information of the dialog is recorded by using a log system and a mysql database.
Further, constructing a query sentence template for the geriatric knowledge graph according to the user input and the intention analysis result, including:
and constructing a query statement template aiming at the elderly medical knowledge graph according to the inquiry template and the determined inquiry intention.
Further, the query sentence template is utilized to query and output the answer in the elderly medical knowledge graph, which comprises the following steps:
inquiring the text information of the answer in the elderly medical knowledge graph by using an inquiry statement template;
And calling a text-to-speech service to convert the text information into speech information.
Correspondingly, the invention also discloses a medical question-answering service system for the elderly, which comprises the following steps:
The map construction module is configured to construct a geriatric knowledge map by integrating the medical database, the clinical guideline and the geriatric study data;
The model training module is configured to construct a data set aiming at the language characteristics and medical requirements of the old and train a natural voice understanding model;
the entity extraction association module is configured to analyze user input through the natural voice understanding model to determine query intention, extract related entities and attributes, and correspond the extracted entities to the old-year medical knowledge graph;
A dialogue management module configured to maintain dialogue history through a finite state machine to ensure consistency of continuous questions and answers;
The template construction module is configured to construct a query sentence template aiming at the elderly medical knowledge graph according to user input and intention analysis results;
And the answer generation module is configured to respond to the input of the acquired user, identify a corresponding query statement template according to the input information, and query and output the answer in the elderly medical knowledge graph by utilizing the query statement template.
Correspondingly, the invention discloses a medical question-answering service device for the elderly, which comprises:
The memory is used for storing a medical question-answer service program for the elderly;
A processor for implementing the steps of the elderly-oriented medical question-answer service method as described in any of the above when executing the elderly-oriented medical question-answer service program.
Accordingly, the invention discloses a readable storage medium, on which a medical question-and-answer service program for the elderly is stored, which when executed by a processor, implements the steps of the medical question-and-answer service method for the elderly as described in any one of the above.
Compared with the prior art, the invention has the beneficial effects that: the invention discloses a medical question-answering service method, a system, a device and a medium for the elderly, which adopt a mode of constructing a special elderly medical knowledge graph and combine deep learning-based natural language understanding, dialogue management and graph query sentence templates to generate corresponding answers, thereby solving the problem that the questions and answers of the specific needs of the elderly cannot be met under the scene based on a general knowledge graph.
According to the invention, the special expression and language habit of the old are captured, and the model optimization is carried out according to the special requirements and language characteristics of the old, so that the complex medical questions and answers of the old can be accurately dealt with, and the accuracy of the medical questions and answers is effectively improved.
It can be seen that the present invention has outstanding substantial features and significant advances over the prior art, as well as the benefits of its implementation.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings that are required to be used in the embodiments or the description of the prior art will be briefly described below, and it is obvious that the drawings in the following description are only embodiments of the present invention, and that other drawings can be obtained according to the provided drawings without inventive effort for a person skilled in the art.
FIG. 1 is a flow chart of a method of an embodiment of the present invention;
Fig. 2 is a system configuration diagram of an embodiment of the present invention.
In the figure, 1, a map construction module; 2. a model training module; 3. entity extraction association module; 4. a dialogue management module; 5. a template construction module; 6. and an answer generation module.
Detailed Description
In order to better understand the aspects of the present invention, the present invention will be described in further detail with reference to the accompanying drawings and detailed description. It will be apparent that the described embodiments are only some, but not all, embodiments of the invention. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
Embodiment one:
As shown in fig. 1, the embodiment provides a medical question-answering service method for the elderly, which includes the following steps:
s1: and constructing an elderly medical knowledge graph by integrating a medical database, clinical guidelines and elderly medical research data.
In a specific embodiment, the specific flow of this step is as follows:
1. Medical data from medical literature, medical encyclopedia, clinical guidelines, and geriatric medical research data is collected, and the medical vocabulary therein is extracted and summarized using jieba word segmentation tools.
2. Randomly sampling the extracted vocabulary content, taking 10% of samples to label entity types, and determining the relation among entities.
3. Medical data from medical literature, medical encyclopedia, clinical guidelines, and geriatric medical research data are determined and integrated for data cleansing and preprocessing to ensure unification of data formats.
4. And extracting the entity and the relation between the entity and the relation by using the UIE to integrate the cleaned data.
5. And verifying the relationship between the entities, and generating an elderly medical knowledge graph after the verification is passed.
As an example, the aim of this step is to construct a geriatric knowledge graph. Including determining key entities in the map, such as diseases, symptoms, drugs, etc., defining relationships between the entities, and integrating medical databases, clinical guidelines, and geriatric medical studies into the map.
The specific flow includes determining entity category and relationship category, data integration and extracting entity-to-entity relationship.
Determining entity category and relation category: first, document analysis is performed, and authoritative medical data from medical documents, medical encyclopedias, clinical guidelines, and geriatric research data are collected, and jieba word segmentation tools are used to extract and aggregate words from a large number of medical documents. Then, randomly sampling the extracted vocabulary content, and taking 10% of samples for marking entity types by medical staff. And cooperates with the expert in the elderly medical field to ensure that all important entity categories are already contained and to determine the relationships between entities that may exist.
Data integration: authoritative medical data from medical literature, medical encyclopedia, clinical guidelines, and geriatric medical research data is first determined and integrated. Then, before integrating the data, data is cleaned and preprocessed to ensure the unification of the data formats, thereby facilitating model extraction.
Extracting the relation between entities: and firstly, extracting the entity and the relation between the entity by using the UIE. The accuracy and integrity of the extracted entities and relationships between the entities is then ensured again by medical expert verification.
S2: a data set is constructed according to the language characteristics and medical requirements of the old, and a natural voice understanding model is trained.
In a specific embodiment, a data set containing data of an elderly person's medical consultation is first collected and organized. Then, analyzing the data set, and determining a problem intention template of the old people in the inquiry according to a preset matching mechanism. Wherein the question intent template records the entity category and the user intent. After the analysis is completed, the elderly medical inquiry data is collected and the entities and inquiry intents are marked. And finally, training the BERT-based intent recognition and entity recognition joint model by using the labeling result.
As an example, the purpose of this step is to fine tune the natural language understanding model using deep learning. Among them, it is necessary to optimize particularly for the linguistic features and medical needs of the elderly.
The optimization training process of the model specifically comprises the following steps:
Data acquisition and analysis: a data set containing medical inquiry information of the elderly is first collected and organized in cooperation with a nursing home and a hospital. The collected interview data set is then analyzed in conjunction with a medical professional to determine several question intentions and templates that may occur to the elderly at the time of the interview, wherein the question templates will define in detail the combination of different entity categories and user intentions.
Query intent recognition and entity recognition: firstly, the collected data is transmitted to medical staff for marking, and the entity and the query intention in the data are marked. The labeling results are then used to train a BERT-based joint model of intent recognition and entity recognition.
S3: user input is analyzed through a natural speech understanding model to determine query intent and extract relevant entities and attributes, and the extracted entities are mapped to the geriatric knowledge graph.
In a specific embodiment, first, the old people medical inquiry data input by a user is analyzed through a BERT-based intention recognition and entity recognition combined model, inquiry intention is determined, and related entities and attributes are extracted. After the entity is identified, the extracted entity is linked with the corresponding entity in the elderly medical knowledge graph by using a medical term standardization technology.
S4: the dialogue history is maintained by a finite state machine to ensure consistency of continuous questions and answers.
In the specific embodiment, firstly, states and transition conditions of the states at different stages of a dialogue are defined; then adding an exception handling mechanism in the finite state machine to handle exceptions or errors occurring in the conversation process; finally, state tracking is carried out, and the state change and conversion information of the dialogue are recorded by using a log system and a mysql database.
Illustratively, this step ensures consistency of the continuous questions and answers by a Finite State Machine (FSM) maintaining a dialogue history. The specific implementation mode is as follows:
State definition: a series of states are defined to represent different phases of a conversation, such as "inquiry Start", "inquiry health information state", "inquiry medicine information state", etc
Defining a trigger: a series of transition rules are formulated to determine the transition conditions from one state to another.
Exception handling: an exception handling mechanism is added to the FSM to handle exceptions or errors that may occur during the conversation, ensuring that the conversation can proceed smoothly.
State tracking and logging: state tracking is performed, and state changes and transitions of conversations are recorded by using a log system and a mysql database.
S5: and constructing a query statement template aiming at the elderly medical knowledge graph according to the user input and the intention analysis result.
In a specific embodiment, a query sentence template for the geriatric knowledge graph is constructed according to the query templates and the determined query intentions.
As an example, a query sentence template for the geriatric knowledge graph is constructed according to the previously determined inquiry template and the identified entities and intentions input by the user, thereby querying the answer in the geriatric knowledge graph.
S6: and after the user input is obtained, identifying a corresponding query statement template according to the input information, and querying and outputting an answer in the elderly medical knowledge graph by utilizing the query statement template.
In a specific embodiment, according to the intention of a user, the content of the atlas query is sleeved in a corresponding query statement template, and the text information of the answer is queried in the senile medical knowledge atlas by utilizing the query statement template; and calling a text-to-speech service to convert the text information into speech information.
The embodiment provides a medical question-answering service method for the elderly, which can accurately cope with complex medical question-answering of the elderly by capturing specific expression and language habits of the elderly and optimizing a model according to special requirements and language characteristics of the elderly, and effectively improves the accuracy of medical question-answering.
Embodiment two:
based on the first embodiment, as shown in fig. 2, the invention also discloses a medical question-answering service system for the elderly, which comprises the following steps: the system comprises a map construction module 1, a model training module 2, an entity extraction association module 3, a dialogue management module 4, a template construction module 5 and an answer generation module 6.
The atlas construction module 1 is configured to construct an elderly medical knowledge atlas by integrating a medical database, clinical guidelines and elderly medical research data.
As an example, the map construction module 1 is specifically for: collecting medical data from medical literature, medical encyclopedia, clinical guidelines and geriatric medical research data, and extracting and summarizing medical vocabulary therein using jieba word segmentation tools; randomly sampling the extracted vocabulary content, taking 10% of samples for labeling entity types, and determining the relationship among entities; determining and integrating medical data from medical literature, medical encyclopedia, clinical guidelines and geriatric medical research data, and performing data cleaning and preprocessing to ensure the unification of data formats; using UIE to extract entities and relations between the entities and the entities; and verifying the relationship between the entities, and generating an elderly medical knowledge graph after the verification is passed.
The model training module 2 is configured to construct a data set for language features and medical needs of the elderly and train a natural speech understanding model.
As an example, the model training module 2 is specifically configured to: collecting and organizing a data set containing medical consultation data of the elderly; analyzing a data set, and determining a problem intention template of the elderly during inquiry according to a preset matching mechanism, wherein the problem intention template comprises entity types and user intentions; collecting medical inquiry data of the old, and labeling entities and inquiry intentions in the medical inquiry data;
and training a BERT-based intent recognition and entity recognition joint model by using the labeling result.
An entity extraction association module 3 configured to analyze the user input through a natural speech understanding model to determine query intent and extract relevant entities and attributes, and to correspond the extracted entities into a geriatric knowledge graph.
As an example, the entity extraction association module 3 is specifically configured to: analyzing senile medical inquiry data input by a user through a BERT-based intention recognition and entity recognition combined model, determining inquiry intention and extracting related entities and attributes; and linking the extracted entity with the corresponding entity in the elderly medical knowledge graph by using a medical term standardization technology.
The dialogue management module 4 is configured to maintain dialogue history through a finite state machine to ensure consistency of continuous questions and answers.
As an example, the dialog management module 4 is specifically configured to: defining states and transition conditions of the states at different stages of the dialogue; adding an exception handling mechanism in the finite state machine to handle exceptions or errors occurring during the conversation; state tracking is performed and state change and transition information of the dialog is recorded by using a log system and a mysql database.
The template construction module 5 is configured to construct a query sentence template for the geriatric knowledge graph according to user input and intention analysis results.
As an example, the template construction module 5 is specifically configured to: and constructing a query statement template aiming at the elderly medical knowledge graph according to the inquiry template and the determined inquiry intention.
And the answer generation module 6 is configured to respond to the input of the user, identify a corresponding query statement template according to the input information, and query and output the answer in the elderly medical knowledge graph by utilizing the query statement template.
Embodiment III:
The embodiment discloses a medical question-answering service device for the elderly, which comprises a processor and a memory; the processor realizes the following steps when executing the medical question-answering service program for the elderly stored in the memory:
1. And constructing an elderly medical knowledge graph by integrating a medical database, clinical guidelines and elderly medical research data.
2. A data set is constructed according to the language characteristics and medical requirements of the old, and a natural voice understanding model is trained.
3. User input is analyzed through a natural speech understanding model to determine query intent and extract relevant entities and attributes, and the extracted entities are mapped to the geriatric knowledge graph.
4. The dialogue history is maintained by a finite state machine to ensure consistency of continuous questions and answers.
5. And constructing a query statement template aiming at the elderly medical knowledge graph according to the user input and the intention analysis result.
6. And after the user input is obtained, identifying a corresponding query statement template according to the input information, and querying and outputting an answer in the elderly medical knowledge graph by utilizing the query statement template.
Further, the medical question-answering service device for the elderly in this embodiment may further include:
The input interface is used for acquiring an externally imported medical question-and-answer service program facing the elderly, storing the acquired medical question-and-answer service program facing the elderly in the memory, and acquiring various instructions and parameters transmitted by external terminal equipment and transmitting the instructions and parameters to the processor so that the processor can develop corresponding processing by utilizing the various instructions and parameters. In this embodiment, the input interface may specifically include, but is not limited to, a USB interface, a serial interface, a voice input interface, a fingerprint input interface, a hard disk reading interface, and the like.
And the output interface is used for outputting various data generated by the processor to the terminal equipment connected with the output interface so that other terminal equipment connected with the output interface can acquire various data generated by the processor. In this embodiment, the output interface may specifically include, but is not limited to, a USB interface, a serial interface, and the like.
And the communication unit is used for establishing remote communication connection between the medical question-answering service device facing the elderly and the external server so that the medical question-answering service device facing the elderly can mount the mirror image file to the external server. In this embodiment, the communication unit may specifically include, but is not limited to, a remote communication unit based on a wireless communication technology or a wired communication technology.
And the keyboard is used for acquiring various parameter data or instructions input by a user by knocking the key cap in real time.
And the display is used for running the relevant information of the medical question-answering service process for the elderly to display in real time.
A mouse may be used to assist a user in inputting data and to simplify user operations.
Embodiment four:
The present embodiment also discloses a readable storage medium, where the readable storage medium includes Random Access Memory (RAM), memory, read Only Memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, registers, hard disk, removable disk, CD-ROM, or any other form of storage medium known in the art. The readable storage medium stores a medical question-answer service program for the elderly, and the medical question-answer service program for the elderly realizes the following steps when being executed by a processor:
1. And constructing an elderly medical knowledge graph by integrating a medical database, clinical guidelines and elderly medical research data.
2. A data set is constructed according to the language characteristics and medical requirements of the old, and a natural voice understanding model is trained.
3. User input is analyzed through a natural speech understanding model to determine query intent and extract relevant entities and attributes, and the extracted entities are mapped to the geriatric knowledge graph.
4. The dialogue history is maintained by a finite state machine to ensure consistency of continuous questions and answers.
5. And constructing a query statement template aiming at the elderly medical knowledge graph according to the user input and the intention analysis result.
6. And after the user input is obtained, identifying a corresponding query statement template according to the input information, and querying and outputting an answer in the elderly medical knowledge graph by utilizing the query statement template.
In summary, the invention combines deep learning-based natural language understanding, dialogue management and map query by constructing the special senile medical knowledge map so as to generate an answer, thereby effectively improving the accuracy of medical questions and answers.
In this specification, each embodiment is described in a progressive manner, and each embodiment is mainly described in a different point from other embodiments, so that the same or similar parts between the embodiments are referred to each other. For the method disclosed in the embodiment, since it corresponds to the system disclosed in the embodiment, the description is relatively simple, and the relevant points refer to the description of the method section.
Those of skill would further appreciate that the various illustrative elements and algorithm steps described in connection with the embodiments disclosed herein may be implemented as electronic hardware, computer software, or combinations of both, and that the various illustrative elements and steps are described above generally in terms of functionality in order to clearly illustrate the interchangeability of hardware and software. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the solution. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present invention.
In the several embodiments provided by the present invention, it should be understood that the disclosed systems, and methods may be implemented in other ways. For example, the system embodiments described above are merely illustrative, e.g., the division of the elements is merely a logical functional division, and there may be additional divisions when actually implemented, e.g., multiple elements or components may be combined or integrated into another system, or some features may be omitted or not performed. Alternatively, the coupling or direct coupling or communication connection shown or discussed with each other may be through some interface, system or unit indirect coupling or communication connection, which may be in electrical, mechanical or other form.
The units described as separate units may or may not be physically separate, and units shown as units 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 units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.
In addition, each functional module in the embodiments of the present invention may be integrated in one processing unit, or each module may exist alone physically, or two or more modules may be integrated in one unit.
Similarly, each processing unit in the embodiments of the present invention may be integrated in one functional module, or each processing unit may exist physically, or two or more processing units may be integrated in one functional module.
The steps of a method or algorithm described in connection with the embodiments disclosed herein may be embodied directly in hardware, in a software module executed by a processor, or in a combination of the two. The software modules may be disposed in Random Access Memory (RAM), memory, read Only Memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, registers, hard disk, a removable disk, a CD-ROM, or any other form of storage medium known in the art.
Finally, it is further noted that relational terms such as first and second, and the like are 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. Moreover, 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 one … …" does not exclude the presence of other like elements in a process, method, article, or apparatus that comprises the element.
The medical question-answering service method, the system and the device for the elderly people, provided by the invention, are described in detail. The principles and embodiments of the present invention have been described herein with reference to specific examples, the description of which is intended only to facilitate an understanding of the method of the present invention and its core ideas. It should be noted that it will be apparent to those skilled in the art that various modifications and adaptations of the invention can be made without departing from the principles of the invention and these modifications and adaptations are intended to be within the scope of the invention as defined in the following claims.
Claims (10)
1. A medical question-answering service method for elderly people, comprising:
Constructing an elderly medical knowledge graph by integrating a medical database, clinical guidelines and elderly medical research data;
constructing a data set aiming at the language characteristics and medical requirements of the old and training a natural voice understanding model;
Analyzing user input through a natural voice understanding model to determine query intention, extracting related entities and attributes, and corresponding the extracted entities to the old-fashioned medical knowledge graph;
Maintaining a dialogue history through a finite state machine to ensure consistency of continuous questions and answers;
Constructing a query sentence template aiming at the elderly medical knowledge graph according to user input and intention analysis results;
In response to obtaining user input, identifying a corresponding query sentence template according to the input information,
And inquiring the answers in the elderly medical knowledge graph by using the inquiry statement template and outputting the answers.
2. The senile-oriented medical question-answering service method according to claim 1, wherein the construction of the senile medical knowledge graph by integrating a medical database, a clinical guideline and a senile medical study data comprises:
collecting medical data from medical literature, medical encyclopedia, clinical guidelines and geriatric medical research data, and extracting and summarizing medical vocabulary therein using jieba word segmentation tools;
Randomly sampling the extracted vocabulary content, taking 10% of samples for labeling entity types, and determining the relationship among entities;
determining and integrating medical data from medical literature, medical encyclopedia, clinical guidelines and geriatric medical research data, and performing data cleaning and preprocessing to ensure the unification of data formats;
using UIE to extract entities and relations between the entities and the entities;
And verifying the relationship between the entities, and generating an elderly medical knowledge graph after the verification is passed.
3. The senile-oriented medical question-answering service method according to claim 1, wherein the constructing a data set for language features and medical needs of the elderly and training a natural speech understanding model comprises:
collecting and organizing a data set containing medical consultation data of the elderly;
analyzing a data set, and determining a problem intention template of the elderly during inquiry according to a preset matching mechanism, wherein the problem intention template comprises entity types and user intentions;
Collecting medical inquiry data of the old, and labeling entities and inquiry intentions in the medical inquiry data;
and training a BERT-based intent recognition and entity recognition joint model by using the labeling result.
4. The elderly-oriented medical question-answering service method according to claim 3, wherein analyzing user input through a natural speech understanding model to determine query intention and extract related entities and attributes, and mapping the extracted entities into an elderly medical knowledge-graph, comprises:
Analyzing senile medical inquiry data input by a user through a BERT-based intention recognition and entity recognition combined model, determining inquiry intention and extracting related entities and attributes;
and linking the extracted entity with the corresponding entity in the elderly medical knowledge graph by using a medical term standardization technology.
5. The elderly-oriented medical question-answering service method according to claim 4, wherein the maintaining of dialogue history by a finite state machine to ensure consistency of continuous questions-answers comprises:
Defining states and transition conditions of the states at different stages of the dialogue;
Adding an exception handling mechanism in the finite state machine to handle exceptions or errors occurring during the conversation;
State tracking is performed and state change and transition information of the dialog is recorded by using a log system and a mysql database.
6. The elderly-oriented medical question-answering service method according to claim 5, wherein the constructing a query sentence template for an elderly medical knowledge graph according to user input and intention analysis results comprises:
and constructing a query statement template aiming at the elderly medical knowledge graph according to the inquiry template and the determined inquiry intention.
7. The elderly-oriented medical question-answering service method according to claim 1, wherein the query sentence template is used to query and output answers in an elderly medical knowledge graph, comprising:
inquiring the text information of the answer in the elderly medical knowledge graph by using an inquiry statement template;
And calling a text-to-speech service to convert the text information into speech information.
8. An elderly-oriented medical question-answering service system, comprising:
The map construction module is configured to construct a geriatric knowledge map by integrating the medical database, the clinical guideline and the geriatric study data;
The model training module is configured to construct a data set aiming at the language characteristics and medical requirements of the old and train a natural voice understanding model;
the entity extraction association module is configured to analyze user input through the natural voice understanding model to determine query intention, extract related entities and attributes, and correspond the extracted entities to the old-year medical knowledge graph;
A dialogue management module configured to maintain dialogue history through a finite state machine to ensure consistency of continuous questions and answers;
The template construction module is configured to construct a query sentence template aiming at the elderly medical knowledge graph according to user input and intention analysis results;
An answer generation module configured to identify a corresponding query sentence template according to input information in response to the user input,
And inquiring the answers in the elderly medical knowledge graph by using the inquiry statement template and outputting the answers.
9. A medical question-answering service device for elderly people, comprising:
The memory is used for storing a medical question-answer service program for the elderly;
a processor for implementing the steps of the elderly-oriented medical question-answering service method according to any one of claims 1 to 7 when executing the elderly-oriented medical question-answering service program.
10. A readable storage medium, characterized by: the readable storage medium has stored thereon an elderly-oriented medical question-and-answer service program which, when executed by a processor, implements the steps of the elderly-oriented medical question-and-answer service method of any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410317346.8A CN117909487A (en) | 2024-03-20 | 2024-03-20 | Medical question-answering service method, system, device and medium for old people |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410317346.8A CN117909487A (en) | 2024-03-20 | 2024-03-20 | Medical question-answering service method, system, device and medium for old people |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117909487A true CN117909487A (en) | 2024-04-19 |
Family
ID=90686342
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410317346.8A Pending CN117909487A (en) | 2024-03-20 | 2024-03-20 | Medical question-answering service method, system, device and medium for old people |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117909487A (en) |
Citations (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106021281A (en) * | 2016-04-29 | 2016-10-12 | 京东方科技集团股份有限公司 | Method for establishing medical knowledge graph, device for same and query method for same |
| CN106911812A (en) * | 2017-05-05 | 2017-06-30 | 腾讯科技（上海）有限公司 | A kind of processing method of session information, server and computer-readable recording medium |
| CN109471948A (en) * | 2018-11-08 | 2019-03-15 | 威海天鑫现代服务技术研究院有限公司 | A kind of the elder's health domain knowledge question answering system construction method |
| CN109472033A (en) * | 2018-11-19 | 2019-03-15 | 华南师范大学 | Entity relation extraction method and system in text, storage medium, electronic equipment |
| CN111125309A (en) * | 2019-12-23 | 2020-05-08 | 中电云脑(天津)科技有限公司 | Natural language processing method and device, computing equipment and storage medium |
| CN111858888A (en) * | 2020-07-13 | 2020-10-30 | 北京航空航天大学 | Multi-round dialogue system of check-in scene |
| CN112148851A (en) * | 2020-09-09 | 2020-12-29 | 常州大学 | Construction method of medicine knowledge question-answering system based on knowledge graph |
| CN113707303A (en) * | 2021-08-30 | 2021-11-26 | 康键信息技术（深圳）有限公司 | Method, device, equipment and medium for solving medical problems based on knowledge graph |
| CN114203165A (en) * | 2021-11-30 | 2022-03-18 | 南京理工大学 | Method and system for quickly identifying incomplete voice of old people |
| CN115269806A (en) * | 2022-08-08 | 2022-11-01 | 中国地质大学（北京） | Question-answering method, electronic device and storage medium applied to mineral domain knowledge graph |
| CN117251543A (en) * | 2023-09-11 | 2023-12-19 | 国药集团新疆新特药业有限公司 | Knowledge graph-based medicine supply chain question answering method and system |
| CN117313785A (en) * | 2023-10-07 | 2023-12-29 | 山东新一代信息产业技术研究院有限公司 | Intelligent digital human interaction method, device and medium based on weak population |
| CN117591655A (en) * | 2023-12-13 | 2024-02-23 | 湖南中医药大学 | Intelligent question-answering system based on traditional Chinese medicine knowledge graph |
-
2024
- 2024-03-20 CN CN202410317346.8A patent/CN117909487A/en active Pending
Patent Citations (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106021281A (en) * | 2016-04-29 | 2016-10-12 | 京东方科技集团股份有限公司 | Method for establishing medical knowledge graph, device for same and query method for same |
| CN106911812A (en) * | 2017-05-05 | 2017-06-30 | 腾讯科技（上海）有限公司 | A kind of processing method of session information, server and computer-readable recording medium |
| CN109471948A (en) * | 2018-11-08 | 2019-03-15 | 威海天鑫现代服务技术研究院有限公司 | A kind of the elder's health domain knowledge question answering system construction method |
| CN109472033A (en) * | 2018-11-19 | 2019-03-15 | 华南师范大学 | Entity relation extraction method and system in text, storage medium, electronic equipment |
| CN111125309A (en) * | 2019-12-23 | 2020-05-08 | 中电云脑(天津)科技有限公司 | Natural language processing method and device, computing equipment and storage medium |
| CN111858888A (en) * | 2020-07-13 | 2020-10-30 | 北京航空航天大学 | Multi-round dialogue system of check-in scene |
| CN112148851A (en) * | 2020-09-09 | 2020-12-29 | 常州大学 | Construction method of medicine knowledge question-answering system based on knowledge graph |
| CN113707303A (en) * | 2021-08-30 | 2021-11-26 | 康键信息技术（深圳）有限公司 | Method, device, equipment and medium for solving medical problems based on knowledge graph |
| CN114203165A (en) * | 2021-11-30 | 2022-03-18 | 南京理工大学 | Method and system for quickly identifying incomplete voice of old people |
| CN115269806A (en) * | 2022-08-08 | 2022-11-01 | 中国地质大学（北京） | Question-answering method, electronic device and storage medium applied to mineral domain knowledge graph |
| CN117251543A (en) * | 2023-09-11 | 2023-12-19 | 国药集团新疆新特药业有限公司 | Knowledge graph-based medicine supply chain question answering method and system |
| CN117313785A (en) * | 2023-10-07 | 2023-12-29 | 山东新一代信息产业技术研究院有限公司 | Intelligent digital human interaction method, device and medium based on weak population |
| CN117591655A (en) * | 2023-12-13 | 2024-02-23 | 湖南中医药大学 | Intelligent question-answering system based on traditional Chinese medicine knowledge graph |
Non-Patent Citations (1)
| Title |
|---|
| 李梦亭: "《基于知识图谱的老年人膳食问答推荐***的设计与实现》", 《中国优秀硕士学位论文全文数据库 信息科技辑》, 31 December 2021 (2021-12-31) * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111414393B (en) | Semantic similar case retrieval method and equipment based on medical knowledge graph | |
| CN112786194A (en) | Medical image diagnosis guide inspection system, method and equipment based on artificial intelligence | |
| CN112863630A (en) | Personalized accurate medical question-answering system based on data and knowledge | |
| CN107368547A (en) | A kind of intelligent medical automatic question-answering method based on deep learning | |
| CN109920540A (en) | Construction method, device and the computer equipment of assisting in diagnosis and treatment decision system | |
| CN112015917A (en) | Data processing method and device based on knowledge graph and computer equipment | |
| CN112037910B (en) | Health information management method, device, equipment and storage medium | |
| CN111768869B (en) | Medical guide mapping construction search system and method for intelligent question-answering system | |
| CN109378066A (en) | A kind of control method and control device for realizing disease forecasting based on feature vector | |
| CN106251865A (en) | A kind of medical treatment & health record Auto-writing method based on speech recognition | |
| CN110610751A (en) | Electronic medical record inputting system and method | |
| CN112599213B (en) | Classification code determining method, device, equipment and storage medium | |
| CN113569023A (en) | Chinese medicine question-answering system and method based on knowledge graph | |
| CN113764112A (en) | Online medical question and answer method | |
| CN111008269A (en) | Data processing method and device, storage medium and electronic terminal | |
| CN112347257B (en) | Patient symptom oral language standardization method and device | |
| CN113886716B (en) | Emergency disposal recommendation method and system for food safety emergencies | |
| CN113409907A (en) | Intelligent pre-inquiry method and system based on Internet hospital | |
| CN111429985B (en) | Electronic medical record data processing method and system | |
| CN113435200A (en) | Entity recognition model training and electronic medical record processing method, system and equipment | |
| CN117809798A (en) | Verification report interpretation method, system, equipment and medium based on large model | |
| CN112349367B (en) | Method, device, electronic equipment and storage medium for generating simulated medical record | |
| CN117854748A (en) | Knowledge graph and generation-based large model inquiry method and system | |
| CN117909487A (en) | Medical question-answering service method, system, device and medium for old people | |
| CN114328864A (en) | Ophthalmic question-answering system based on artificial intelligence and knowledge graph |
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 |