CN116881413A - Intelligent medical question-answering method based on Chinese medical knowledge - Google Patents
Intelligent medical question-answering method based on Chinese medical knowledge Download PDFInfo
- Publication number
- CN116881413A CN116881413A CN202310773436.3A CN202310773436A CN116881413A CN 116881413 A CN116881413 A CN 116881413A CN 202310773436 A CN202310773436 A CN 202310773436A CN 116881413 A CN116881413 A CN 116881413A
- Authority
- CN
- China
- Prior art keywords
- medical
- user
- question
- chinese medical
- intelligent
- 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 16
- 238000003745 diagnosis Methods 0.000 claims abstract description 12
- 201000010099 disease Diseases 0.000 claims description 5
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 claims description 5
- 239000003814 drug Substances 0.000 claims description 5
- 229940079593 drug Drugs 0.000 claims description 4
- 208000024891 symptom Diseases 0.000 claims description 4
- 238000011282 treatment Methods 0.000 claims description 4
- 238000013473 artificial intelligence Methods 0.000 abstract description 2
- 235000002198 Annona diversifolia Nutrition 0.000 description 2
- 241000282842 Lama glama Species 0.000 description 2
- 238000013459 approach Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000004044 response Effects 0.000 description 1
- 238000013519 translation Methods 0.000 description 1
- 238000011269 treatment regimen Methods 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
- G06F16/33—Querying
- G06F16/3331—Query processing
- G06F16/334—Query execution
- G06F16/3344—Query execution using natural language analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/35—Clustering; Classification
-
- 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
- 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
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
- Y02A90/00—Technologies having an indirect contribution to adaptation to climate change
- Y02A90/10—Information and communication technologies [ICT] supporting adaptation to climate change, e.g. for weather forecasting or climate simulation
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Databases & Information Systems (AREA)
- Data Mining & Analysis (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Computational Linguistics (AREA)
- Health & Medical Sciences (AREA)
- Public Health (AREA)
- Artificial Intelligence (AREA)
- Biomedical Technology (AREA)
- Medical Informatics (AREA)
- Mathematical Physics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Animal Behavior & Ethology (AREA)
- Human Computer Interaction (AREA)
- Pathology (AREA)
- Epidemiology (AREA)
- General Health & Medical Sciences (AREA)
- Primary Health Care (AREA)
- Medical Treatment And Welfare Office Work (AREA)
Abstract
An intelligent medical question-answering method based on Chinese medical knowledge relates to the technical field of artificial intelligence, and is characterized in that a Chinese medical knowledge graph is constructed, relevant medical knowledge is extracted by means of GPT3.5API, and more than 6000 instruction data are generated for supervision and fine adjustment. The model takes LLaMA-7B as a basic model, and the generated instruction data is utilized for fine adjustment, so that the model has rich medical field expertise, and a more specialized answer is made for intelligent diagnosis.
Description
Technical Field
The invention relates to the technical field of artificial intelligence, in particular to an intelligent medical question-answering method based on Chinese medical knowledge.
Background
With the development of society, people have higher and higher demands on medical health, but doctor resources are limited, and the phenomena of difficult and expensive doctor seeing of patients still exist. To address this problem, intelligent medical question-answering systems have evolved. However, most of the intelligent medical question-answering systems on the market at present are mainly English, are not friendly to Chinese users, and lack support of a Chinese medical knowledge base, so that answering accuracy is low and user experience is poor.
The medical field is a very large and complex field, and includes various diseases, symptoms, treatment methods, medicines, and the like. The doctor needs to have a lot of medical knowledge and experience in practice to be able to make the correct diagnosis and treatment regimen. In recent years, medical demands have increased and medical resources have become more and more intense due to population growth and lifestyle changes. Therefore, the intelligent medical question-answering system is established, intelligent consultation and advice can be provided for medical staff and patients, and the intelligent medical question-answering system has very important significance for relieving medical pressure and improving medical efficiency. Because of the great expertise in the medical field, LLMs are often unable to meet the specialized needs in this field, and there are still problems associated with their use in the medical field, whether they be the original LLaMA or ChatGPT, among other large language models. For example, inputting a piece of disease description into LLaMA, and letting it output disease diagnosis information, it gives some very brief and routine answers, sometimes even no answer at all, if it is directly used for intelligent diagnosis, it is very likely to cause unscientific in terms of diagnosis accuracy, medicine recommendation, medical advice, etc., and even endanger the life of the patient. Therefore, it is necessary to input the diagnosis case data to a large model for specialized learning with specialized medical field knowledge. Currently, there have been some approaches to try to solve this problem, but these approaches rely mainly on retrieving medical information from manual communication, which is prone to human error. Moreover, LLMs are typically trained only in the english context, which limits their understanding and response capabilities in other language environments, such as chinese, and therefore their use in the chinese context is greatly limited. The existing method mainly adopts the ChatGPT for data assistance, and effectively distills the knowledge of the ChatGPT in a certain field to a smaller model: for example, to solve the problem of Chinese context, the DoctorGLM uses ChatGLM-6B as a base model and fine-tunes with the Chinese translation of the ChatDoctor dataset through ChatGPT retrieval. The effect of these models, although improved over the original models, is far from truly landing.
Disclosure of Invention
In order to overcome the defects of the technology, the invention provides a method which has rich expertise in the medical field, so as to make a more specialized answer for intelligent diagnosis.
The technical scheme adopted for overcoming the technical problems is as follows:
an intelligent medical question-answering method based on Chinese medical knowledge comprises the following steps:
s01, classifying the problems of the user;
s02, analyzing the semantics of the user question;
s03, generating natural language answers by using the trimmed model;
s04, recommending relevant medical knowledge and medical services to the user according to the problems and the history record information which are presented by the user.
Further, in step S01, the problems of the user are classified into four categories, namely, symptoms, diagnosis, treatment method, and drug consultation.
Further, in step S03, a Chinese medical instruction data set is generated from the Chinese medical knowledge base by using the GPT3.5API, and the LLaMA-7B model is trimmed by using the Chinese medical instruction data set.
The beneficial effects of the invention are as follows: by constructing a Chinese medical knowledge graph, extracting relevant medical knowledge by means of GPT3.5API, generating more than 6000 instruction data for supervision and fine adjustment. The model takes LLaMA-7B as a basic model, and the generated instruction data is utilized for fine adjustment, so that the model has rich medical field expertise, and a more specialized answer is made for intelligent diagnosis.
Drawings
FIG. 1 is a flow chart of the method of the present invention.
Detailed Description
The invention is further described with reference to fig. 1.
An intelligent medical question-answering method based on Chinese medical knowledge comprises the following steps:
s01, classifying the problems of the user in order to better understand the problems of the user. By categorizing the questions, we can better provide accurate answers to the user.
S02, on the basis of problem classification, intention recognition is needed, namely, the purpose and the requirement of user questioning are understood. By analyzing the semantics of the user questions, the user can better understand the intention and give out answers more in line with the user requirements.
S03, generating natural language answers by using the trimmed model.
S04, recommending relevant medical knowledge and medical services to the user according to the problems and the history record information which are presented by the user, and improving user satisfaction and experience. .
By constructing a Chinese medical knowledge graph, extracting relevant medical knowledge by means of GPT3.5API, generating more than 6000 instruction data for supervision and fine adjustment. The model takes LLaMA-7B as a basic model, and the generated instruction data is utilized for fine adjustment, so that the model has rich medical field expertise, and a more specialized answer is made for intelligent diagnosis.
In one embodiment of the present invention, the user' S questions are classified into four categories of symptoms, disease diagnosis, treatment methods, and drug consultation in step S01.
In a specific embodiment of the present invention, in step S03, a Chinese medical instruction dataset is generated for the Chinese medical knowledge base using the GPT3.5API, and the LLaMA-7B model is trimmed using the Chinese medical instruction dataset. Finally, it should be noted that: the foregoing description is only a preferred embodiment of the present invention, and the present invention is not limited thereto, but it is to be understood that modifications and equivalents of some of the technical features described in the foregoing embodiments may be made by those skilled in the art, although the present invention has been described in detail with reference to the foregoing embodiments. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims (3)
1. An intelligent medical question-answering method based on Chinese medical knowledge is characterized by comprising the following steps:
s01, classifying the problems of the user;
s02, analyzing the semantics of the user question;
s03, generating natural language answers by using the trimmed model;
s04, recommending relevant medical knowledge and medical services to the user according to the problems and the history record information which are presented by the user.
2. The intelligent medical question-answering method based on Chinese medical knowledge according to claim 1, wherein: in step S01, the problems of the user are classified into four categories, namely symptoms, disease diagnosis, treatment methods and drug consultation.
3. The intelligent medical question-answering method based on Chinese medical knowledge according to claim 1, wherein: in step S03, a GPT3.5API is used for generating a Chinese medical instruction data set for the Chinese medical knowledge base, and the LLaMA-7B model is subjected to fine tuning by utilizing the Chinese medical instruction data set.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202310773436.3A CN116881413A (en) | 2023-06-28 | 2023-06-28 | Intelligent medical question-answering method based on Chinese medical knowledge |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202310773436.3A CN116881413A (en) | 2023-06-28 | 2023-06-28 | Intelligent medical question-answering method based on Chinese medical knowledge |
Publications (1)
Publication Number | Publication Date |
---|---|
CN116881413A true CN116881413A (en) | 2023-10-13 |
Family
ID=88267048
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202310773436.3A Pending CN116881413A (en) | 2023-06-28 | 2023-06-28 | Intelligent medical question-answering method based on Chinese medical knowledge |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN116881413A (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN117709441A (en) * | 2024-02-06 | 2024-03-15 | 云南联合视觉科技有限公司 | Method for training professional medical large model through gradual migration field |
-
2023
- 2023-06-28 CN CN202310773436.3A patent/CN116881413A/en active Pending
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN117709441A (en) * | 2024-02-06 | 2024-03-15 | 云南联合视觉科技有限公司 | Method for training professional medical large model through gradual migration field |
CN117709441B (en) * | 2024-02-06 | 2024-05-03 | 云南联合视觉科技有限公司 | Method for training professional medical large model through gradual migration field |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN107247868B (en) | Artificial intelligence auxiliary inquiry system | |
CN113871003B (en) | Disease auxiliary differential diagnosis system based on causal medical knowledge graph | |
US11132361B2 (en) | System for responding to complex user input queries using a natural language interface to database | |
JP4615629B2 (en) | Computer-based medical diagnosis and processing advisory system, including access to the network | |
CN112802575B (en) | Medication decision support method, device, equipment and medium based on graphic state machine | |
CN112863630A (en) | Personalized accurate medical question-answering system based on data and knowledge | |
CN110675944A (en) | Triage method and device, computer equipment and medium | |
CN109670179A (en) | Case history text based on iteration expansion convolutional neural networks names entity recognition method | |
KR102424085B1 (en) | Machine-assisted conversation system and medical condition inquiry device and method | |
US20210287800A1 (en) | Ai supported personalized, natural language-based patient interface for medical-bot | |
WO2023029506A1 (en) | Illness state analysis method and apparatus, electronic device, and storage medium | |
CN116881413A (en) | Intelligent medical question-answering method based on Chinese medical knowledge | |
CN115858886B (en) | Data processing method, device, equipment and readable storage medium | |
WO2021143098A1 (en) | Method, apparatus, electronic device, and storage medium for automatically adjusting user priority | |
CN111477320A (en) | Construction system of treatment effect prediction model, treatment effect prediction system and terminal | |
CN116910172A (en) | Follow-up table generation method and system based on artificial intelligence | |
CN116910105A (en) | Medical information query system and method based on pre-training large model | |
Liao et al. | Medical data inquiry using a question answering model | |
CN117216209A (en) | Ultrasonic examination report reading system based on large language model | |
CN111128388A (en) | Value domain data matching method and device and related products | |
EP3901875A1 (en) | Topic modelling of short medical inquiries | |
CN116910213A (en) | Automatic question-answering system based on deep reinforcement learning | |
CN116956934A (en) | Task processing method, device, equipment and storage medium | |
CN115062628A (en) | Automatic simulation method for doctor-patient communication conversation based on knowledge graph | |
CN115565655A (en) | Enhanced auxiliary inquiry method |
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 |