CN115019950A - Epilepsia detection method, system and equipment based on knowledge graph and electroencephalogram - Google Patents
Epilepsia detection method, system and equipment based on knowledge graph and electroencephalogram Download PDFInfo
- Publication number
- CN115019950A CN115019950A CN202210390585.7A CN202210390585A CN115019950A CN 115019950 A CN115019950 A CN 115019950A CN 202210390585 A CN202210390585 A CN 202210390585A CN 115019950 A CN115019950 A CN 115019950A
- Authority
- CN
- China
- Prior art keywords
- epilepsy
- electroencephalogram
- information
- knowledge
- patient
- 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
- G16H50/00—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics
- G16H50/20—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for computer-aided diagnosis, e.g. based on medical expert systems
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/36—Creation of semantic tools, e.g. ontology or thesauri
- G06F16/367—Ontology
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
-
- 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
- G16H50/30—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for calculating health indices; for individual health risk assessment
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Medical Informatics (AREA)
- Health & Medical Sciences (AREA)
- Public Health (AREA)
- Data Mining & Analysis (AREA)
- Biomedical Technology (AREA)
- Databases & Information Systems (AREA)
- Theoretical Computer Science (AREA)
- Primary Health Care (AREA)
- General Health & Medical Sciences (AREA)
- Epidemiology (AREA)
- Pathology (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Software Systems (AREA)
- Animal Behavior & Ethology (AREA)
- Evolutionary Computation (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Computational Linguistics (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Mathematical Physics (AREA)
- Computing Systems (AREA)
- Measurement And Recording Of Electrical Phenomena And Electrical Characteristics Of The Living Body (AREA)
Abstract
The invention discloses an epilepsy detecting method, a system and equipment based on a knowledge-graph and an electroencephalogram, relating to the technical field of intelligent medical treatment, wherein the epilepsy detecting method comprises the steps of obtaining epilepsy medical data and preprocessing the epilepsy medical data; establishing an epileptic disease knowledge map according to the preprocessed epileptic disease medical data; acquiring basic information of a patient, and reasoning the basic information according to an epilepsy knowledge graph to obtain a text vector; acquiring electroencephalogram information of a patient, and inputting the electroencephalogram information into a pre-constructed feature extraction model to obtain an electroencephalogram vector; fusing the text vector and the electroencephalogram image vector to obtain an information characteristic vector of the patient; and inputting the information characteristic vector into a pre-constructed classification model to obtain the epilepsy detection result of the patient. The invention can solve the technical problem of poor detection and identification effect of epilepsy assisted by a single-mode technology in the prior art.
Description
Technical Field
The invention relates to the technical field of intelligent medical treatment, in particular to a epilepsy detection method, system and equipment based on a knowledge graph and electroencephalogram.
Background
With the development of artificial intelligence and big data technology, the artificial intelligence technology plays an increasingly important role in daily life and national economy, the artificial intelligence technology is applied to various fields of production and life, intelligent medical treatment is one of important branches in the artificial intelligence technology, the application aspects of the intelligent medical treatment comprise auxiliary inquiry, auxiliary detection, diagnosis quality control, treatment scheme recommendation and the like, and the intelligent medical treatment technology plays an important role in various links of medical detection. Epilepsy, commonly known as epilepsy or epilepsy, has become the second most common disease in the neurology department in China, and is a chronic disease of transient cerebral dysfunction caused by sudden abnormal discharge of cerebral neurons. The detection of the epilepsy assisted by the artificial intelligence technology is mainly judged by depending on the expert experience of a doctor, and the auxiliary detection system is judged by depending on a rule matching and machine learning method.
At present, a relatively common artificial intelligence technology is mainly used for diagnosing epilepsy, namely, only one characteristic can be identified, in the process of epilepsy detection, basic information of a patient mainly comprises two parts of contents, one part is the basic information and the physical sign information of the patient, characters are used as carriers for recording, the other part is electroencephalogram information of the head of the patient, the electroencephalogram information is reflected in an image form, the electroencephalogram information has an important function in epilepsy detection and is used as an important basis of the epilepsy detection, when the single mode technology is used for processing, only one characteristic can be identified, the two kinds of information cannot be well fused, and the identification effect is poor.
Therefore, the existing epilepsy detection method based on the knowledge graph and the electroencephalogram generally has the technical problem of poor epilepsy detection and identification effects assisted by the monomodal technology.
Disclosure of Invention
Aiming at the defects of the prior art, the invention aims to provide a epilepsy detection method, system and equipment based on a knowledge graph and an electroencephalogram, and aims to solve the technical problem that the effect of assisting epilepsy detection and identification by a single-mode technology in the prior art is poor.
One aspect of the present invention is to provide an epilepsy detection method based on a knowledge-graph and an electroencephalogram, including:
acquiring epilepsy medical data, and preprocessing the epilepsy medical data;
establishing an epileptic disease knowledge map according to the preprocessed epileptic disease medical data;
acquiring basic information of a patient, and reasoning the basic information according to the epilepsy knowledge graph to acquire a text vector;
acquiring electroencephalogram information of the patient, and inputting the electroencephalogram information into a pre-constructed feature extraction model to obtain an electroencephalogram vector;
fusing the text vector and the electroencephalogram image vector to obtain an information feature vector of the patient;
and inputting the information characteristic vector into a pre-constructed classification model to obtain an epilepsy detection result of the patient.
Compared with the prior art, the invention has the beneficial effects that: according to the epilepsy detecting method based on the knowledge graph and the electroencephalogram, a multi-mode technology is adopted for assisting in detecting epilepsy, the problem that an artificial intelligence technology adopts a single-mode technology, only one characteristic can be recognized, and basic information, sign information and head electroencephalogram information of a patient cannot be well fused, so that the effect of assisting in detecting and recognizing the epilepsy is poor is avoided, specifically, epilepsy medical data are obtained, an epilepsy knowledge graph is established according to the epilepsy medical data, and the basic information is reasoned according to the epilepsy knowledge graph to obtain text vectors based on the basic information of the patient; meanwhile, on the basis of electroencephalogram information, electroencephalogram vectors are obtained through a pre-constructed feature extraction model, finally, text vectors and electroencephalogram vectors are fused to obtain information feature vectors of patients, the information feature vectors of the patients are input into a pre-constructed classification model to obtain epileptic detection results of the patients, the intelligent level of artificial intelligent auxiliary detection is improved, the workload of medical workers is relieved, the accuracy of patient detection is improved, and therefore the technical problem that the common single-mode technology auxiliary epileptic detection recognition effect is poor is solved.
According to an aspect of the above technical solution, the step of establishing an epilepsy knowledge map according to the preprocessed epilepsy medical data specifically includes:
performing knowledge extraction on the preprocessed epilepsy medical data to obtain initial ternary group data, wherein the knowledge extraction comprises entity extraction, attribute extraction and relationship extraction;
after knowledge fusion is carried out on the initial triple data, quality evaluation is carried out on the initial triple data to obtain final triple data with qualified quality,
and establishing an epilepsy knowledge graph according to the final ternary group data.
According to an aspect of the above technical solution, the step of obtaining the basic information of the patient and reasoning the basic information according to the epilepsy knowledge graph to obtain the text vector specifically includes:
performing inquiry observation on the patient to acquire basic information of the patient;
and reasoning the basic information according to the epilepsy knowledge graph to obtain a text vector.
According to one aspect of the above technical solution, the step of obtaining the electroencephalogram information of the patient and inputting the electroencephalogram information into a pre-constructed feature extraction model to obtain an electroencephalogram image vector specifically includes:
acquiring electroencephalogram information of a patient through electroencephalogram equipment and preprocessing the electroencephalogram information, wherein the electroencephalogram information comprises image information such as computed tomography, magnetic resonance imaging and electroencephalogram;
constructing an electroencephalogram image feature extraction model;
and inputting the preprocessed electroencephalogram information into the electroencephalogram feature extraction model to obtain an electroencephalogram vector.
According to an aspect of the foregoing technical solution, the step of inputting the information feature vector into a pre-constructed classification model to obtain an epilepsy detection result of the patient specifically includes:
constructing a classification model, and training the classification model according to a plurality of previous cases and diagnosis and treatment records to further obtain an epilepsy classification model;
and inputting the information characteristic vector into the epilepsy classification model to obtain the epilepsy detection result of the patient.
According to one aspect of the above technical solution, the step of constructing the electroencephalogram image feature extraction model specifically includes:
constructing a feature extraction model, wherein the feature extraction model is a DCNN model;
and training the feature extraction model through a plurality of pieces of electroencephalogram image information to further obtain the electroencephalogram image feature extraction model.
According to an aspect of the above technical solution, the step of acquiring epilepsy medical data and preprocessing the epilepsy medical data specifically includes:
acquiring epilepsy medical data, wherein the epilepsy medical data comprises medical data such as medical professional books, hospital electronic diagnosis cases, medical website knowledge and the like;
and preprocessing the epilepsy medical data to remove narrative and descriptive non-knowledge point data contents and retain key part knowledge data.
A second aspect of the present invention provides a system for detecting epilepsy based on a knowledge-graph and electroencephalogram, comprising:
the data preprocessing module is used for acquiring epilepsy medical data and preprocessing the epilepsy medical data;
the knowledge map establishing module is used for establishing an epilepsy knowledge map according to the preprocessed epilepsy medical data;
the text vector output module is used for acquiring basic information of a patient and reasoning the basic information according to the epilepsy knowledge graph to acquire a text vector;
the electroencephalogram image vector acquisition module is used for acquiring electroencephalogram information of the patient and inputting the electroencephalogram information into a pre-constructed feature extraction model to obtain an electroencephalogram vector;
an information feature vector acquisition module for fusing the text vector and the electroencephalogram vector to obtain an information feature vector of the patient;
and the detection result acquisition module is used for inputting the information characteristic vector into a pre-constructed classification model so as to acquire the epilepsy detection result of the patient.
A third aspect of the invention provides a readable storage medium having stored thereon a computer program which, when executed by a processor, carries out the steps of the method as described above.
A fourth aspect of the invention is to provide an electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, the processor implementing the steps of the method described above when executing the program.
Drawings
The above and/or additional aspects and advantages of the present invention will become apparent and readily appreciated from the following description of the embodiments, taken in conjunction with the accompanying drawings of which:
FIG. 1 is a flowchart of a method for detecting epilepsy based on knowledge-graph and electroencephalogram according to a first embodiment of the present invention;
FIG. 2 is a block diagram of a system for epilepsy detection based on a knowledge-graph and electroencephalogram according to a first embodiment of the present invention;
the figure elements are illustrated in symbols:
the system comprises a data preprocessing module 100, a knowledge map establishing module 200, a text vector output module 300, an electroencephalogram image vector obtaining module 400, an information characteristic vector obtaining module 500 and a detection result obtaining module 600.
Detailed Description
In order to make the objects, features and advantages of the present invention comprehensible, embodiments accompanied with figures are described in detail below. Several embodiments of the invention are presented in the drawings. This invention may, however, be embodied in many different forms and should not be construed as limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete.
It will be understood that when an element is referred to as being "secured to" another element, it can be directly on the other element or intervening elements may also be present. When an element is referred to as being "connected" to another element, it can be directly connected to the other element or intervening elements may also be present. As used herein, the terms "vertical," "horizontal," "left," "right," "up," "down," and the like are used for descriptive purposes only and not for purposes of indicating or implying that the referenced device or element must have a particular orientation, be constructed and operated in a particular orientation, and therefore should not be construed as limiting the invention.
In the present invention, unless otherwise expressly specified or limited, the terms "mounted," "connected," "fixed," and the like are to be construed broadly and can, for example, be fixedly connected, detachably connected, or integrally connected; can be mechanically or electrically connected; they may be connected directly or indirectly through intervening media, or they may be interconnected between two elements. The specific meanings of the above terms in the present invention can be understood by those skilled in the art according to specific situations. As used herein, the term "and/or" includes any and all combinations of one or more of the associated listed items.
Example one
Referring to fig. 1, a method for detecting epilepsy based on knowledge mapping and electroencephalogram in multiple modes according to a first embodiment of the present invention is shown, the method including steps S10-S15:
step S10, acquiring epilepsy medical data, and preprocessing the epilepsy medical data;
the epilepsy medical data is obtained and comprises medical data such as medical professional books, hospital electronic diagnosis cases, medical website knowledge and the like. For example, paper data of a professional epilepsy book needs to form an electronic document through an OCR Recognition technology, where the OCR Recognition technology is Optical Character Recognition, i.e. a process in which a computer analyzes and processes an image file of characters and finally obtains a corresponding text file.
Secondly, the epilepsy medical data is preprocessed, namely narrative and descriptive non-knowledge point data contents and the like are removed from the collected epilepsy medical data, irrelevant epilepsy data contents are removed, epilepsy related knowledge data are reserved, and the preprocessed epilepsy medical data serve as a source basis for extracting epilepsy knowledge map data.
Step S11, establishing an epilepsy knowledge map according to the preprocessed epilepsy medical data;
specifically, knowledge extraction is carried out on preprocessed epilepsy medical data to obtain initial ternary group data, wherein the knowledge extraction comprises entity extraction, attribute extraction and relation extraction;
the preprocessed epilepsy medical data comprises structured data, semi-structured data and unstructured data. Structured data refers to data that can be represented and stored in a two-dimensional form using a relational database. Semi-structured data, which is a form of structured data, does not conform to the data model structure associated with relational databases or other forms of data tables, but contains relevant tags to separate semantic elements and to layer records and fields. Unstructured data is data that has no fixed structure. Various documents, pictures, video, audio, etc. belong to unstructured data. Because the semi-structured data and the unstructured data are stored and arranged without rules, great difficulty is brought to the inquiry and modification of the epilepsy medical data.
In order to unify epilepsy medical data, conveniently query data and construct an epilepsy knowledge graph, knowledge extraction needs to be performed on the preprocessed epilepsy medical data, namely, manual and automatic technical processing is performed on semi-structured data and unstructured data, the epilepsy medical data is integrated into unified data and codes, the processing mode comprises entity extraction, attribute extraction and relation extraction, and initial triplet data is formed by combining a regular matching mode.
The entity extraction is to automatically identify a named entity from an original corpus, and as the entity is the most basic element in the epilepsy knowledge graph, the integrity, accuracy, recall rate and the like of the extraction directly influence the quality of the epilepsy knowledge graph. Therefore, entity extraction is the most basic and critical step in knowledge extraction.
After knowledge fusion is carried out on the initial ternary group data, quality evaluation is carried out on the initial ternary group data so as to obtain final ternary group data with qualified quality;
the epilepsy medical data usually exist in a dispersed, heterogeneous and autonomous mode, and have the characteristics of redundancy, noise, uncertainty and incompleteness, knowledge fusion and knowledge processing are required, and the epilepsy medical data of different sources and different structures are fused into a uniform epilepsy knowledge map so as to ensure the uniformity and consistency of knowledge. The knowledge fusion mainly comprises entity disambiguation and reference disambiguation, and it needs to be explained that the entity disambiguation means that one entity reference item can correspond to a plurality of real world entities; simply put, the nature of entity disambiguation is that a word is likely to have multiple meanings, i.e., the meanings indicated in different contexts are not identical. The term "ambiguous" refers to that in linguistics and our commonly used phrases, abbreviation or substitution is used hereinafter to replace a word that has appeared above. Knowledge fusion requires entity disambiguation and reference disambiguation, removes the relative conflicting parts of knowledge redundancy and knowledge, normalizes epileptic medical data, and constructs initial triple data to be more complete and professional.
In addition, the initial triple data with the knowledge fused is subjected to quality evaluation, namely, expert personnel needing to research related specialties of epilepsy carry out quality evaluation on the initial triple data so as to ensure the quality and the speciality of the triple data, and therefore the final triple data with qualified quality is obtained.
And establishing an epilepsy knowledge graph according to the final ternary group data.
And establishing an epileptic disease knowledge map based on the final qualified and unified ternary group data.
Step S12, acquiring basic information of a patient, and reasoning the basic information according to the epilepsy knowledge graph to acquire a text vector;
specifically, inquiring and observing the patient to acquire basic information of the patient;
and reasoning the basic information according to the epilepsy knowledge graph to obtain a text vector.
And reasoning the basic information through the epilepsy knowledge map to obtain a more concise and accurate text vector.
Step S13, acquiring electroencephalogram information of the patient, and inputting the electroencephalogram information into a pre-constructed feature extraction model to acquire an electroencephalogram image vector;
acquiring electroencephalogram information of a patient through electroencephalogram equipment and preprocessing the electroencephalogram information, wherein the electroencephalogram information comprises image information such as computed tomography, magnetic resonance imaging and electroencephalogram;
after the electroencephalogram image of a patient is acquired, preprocessing operations such as cutting and amplifying need to be carried out on the image so as to adapt to an electroencephalogram image feature extraction model.
Constructing an electroencephalogram image feature extraction model;
specifically, a feature extraction model is constructed, wherein the feature extraction model is a DCNN model;
and training the feature extraction model through a plurality of pieces of electroencephalogram image information to further obtain the electroencephalogram image feature extraction model.
And inputting the preprocessed electroencephalogram information into the electroencephalogram feature extraction model to obtain an electroencephalogram vector.
Step S14, fusing the text vector and the electroencephalogram image vector to obtain an information characteristic vector of the patient;
specifically, a text vector, namely an electroencephalogram image vector, is subjected to vector form fusion to obtain an information feature vector of a patient, and the information feature vector stores two information features of the text and the electroencephalogram image.
Step S15, inputting the information feature vector into a pre-constructed classification model to obtain the epilepsy detection result of the patient.
Specifically, a classification model is constructed, and the classification model is trained according to a plurality of previous cases and diagnosis and treatment records, so that an epilepsy classification model is obtained;
and inputting the information characteristic vector into the epilepsy classification model to obtain the epilepsy detection result of the patient.
In the embodiment, text information and electroencephalogram information are fused, the single-mode technology for artificially and intelligently assisting in detecting epilepsy is improved, the intelligent level of artificially and intelligently assisting in detecting epilepsy is improved, the workload of medical workers is relieved, the accuracy of patient diagnosis is improved, misdiagnosis is avoided, the basis is provided for the treatment of next-step epilepsy, meanwhile, the diagnosis level difference among hospitals is reduced by the technology for multimodality artificially and intelligently assisting in detecting epilepsy, the problem of insufficient resources of professional medical treatment and medical workers is solved, the medical cost is saved, and the medical experience of patients is improved.
Compared with the prior art, the epilepsy detection method based on the knowledge graph and the electroencephalogram provided by the embodiment has the beneficial effects that: according to the epilepsy detection method based on the knowledge graph and the electroencephalogram, provided by the invention, the epilepsy is detected in an auxiliary mode by adopting a multi-mode technology, the problem that the artificial intelligence technology adopts a single-mode technology, only one characteristic can be identified, and the basic information, the physical sign information and the head electroencephalogram information of a patient cannot be well fused, so that the auxiliary detection and identification effect of the artificial intelligence epilepsy is poor is avoided, specifically, epilepsy medical data is obtained, the epilepsy knowledge graph is established according to the epilepsy medical data, and the basic information is reasoned according to the epilepsy knowledge graph to obtain a text vector based on the basic information of the patient; meanwhile, on the basis of electroencephalogram information, electroencephalogram vectors are obtained through a pre-constructed feature extraction model, finally, text vectors and electroencephalogram vectors are fused to obtain information feature vectors of patients, the information feature vectors of the patients are input into a pre-constructed classification model to obtain epileptic detection results of the patients, the intelligent level of artificial intelligent auxiliary diagnosis is improved, the workload of medical workers is relieved, the accuracy of patient diagnosis is improved, and therefore the technical problem that the detection and identification effects of the epileptic are poor due to the assistance of a monomodal technology in common is solved.
Example two
Referring to fig. 2, a system for epilepsy detection based on a knowledge-graph and electroencephalogram according to a second embodiment of the present invention is shown, the system includes:
the data preprocessing module 100 is used for acquiring epilepsy medical data and preprocessing the epilepsy medical data;
the epilepsy medical data comprises medical data such as medical professional books, hospital electronic diagnosis cases, medical website knowledge and the like. Collecting and obtaining epilepsy medical data is a source and a basis for constructing epilepsy knowledge graph.
In addition, the epilepsy medical data is preprocessed to remove narrative and descriptive non-knowledge point data contents and keep important part knowledge data. The content of the medical data of unrelated epilepsy is removed, and the medical data related to the epilepsy is reserved so as to obtain more accurate and refined medical data of the epilepsy.
The knowledge map establishing module 200 is used for establishing an epilepsy knowledge map according to the preprocessed epilepsy medical data;
specifically, knowledge extraction is carried out on preprocessed epilepsy medical data to obtain initial ternary group data, wherein the knowledge extraction comprises entity extraction, attribute extraction and relation extraction;
the preprocessed epilepsy medical data comprises structured data, semi-structured data and unstructured data. Because the semi-structured data and the unstructured data are stored and arranged without rules, the inquiry, the check and the modification of the epilepsy medical data are relatively difficult, and therefore the epilepsy medical data needs to be integrated and unified.
In order to unify the epilepsy medical data and facilitate inquiring and checking the epilepsy medical data, knowledge extraction needs to be carried out on the preprocessed epilepsy medical data, namely the epilepsy medical data is subjected to manual and automatic technical processing and integrated into unified data and codes, the processing mode comprises entity extraction, attribute extraction and relation extraction, and initial ternary group data is formed by combining a regular matching mode.
After knowledge fusion is carried out on the initial ternary group data, quality evaluation is carried out on the initial ternary group data so as to obtain final ternary group data with qualified quality;
the epilepsy medical data usually exist in a dispersed, heterogeneous and autonomous mode, and have the characteristics of redundancy, noise, uncertainty and incompleteness, knowledge fusion and knowledge processing are required, and the epilepsy medical data of different sources and different structures are fused into a uniform epilepsy knowledge map so as to ensure the uniformity and consistency of knowledge. The knowledge fusion comprises entity disambiguation and reference disambiguation, the knowledge fusion can remove the knowledge redundancy and the relative knowledge conflict of the epileptic medical data, standardizes and conforms the epileptic medical data, and therefore the initial triple data are constructed into more complete and professional data information.
In addition, the initial triplet group data with the knowledge fused is subjected to quality assessment, namely, experts needing to study the epilepsy related specialties perform quality assessment on the initial triplet group data so as to ensure the quality and the speciality of the triplet group data, and therefore the final triplet group data with qualified quality is obtained.
And establishing an epilepsy knowledge graph according to the final ternary group data.
And establishing an epileptic disease knowledge graph based on the final ternary data with qualified quality and consistency.
A text vector output module 300, configured to obtain basic information of the patient, and perform inference on the basic information according to the epilepsy knowledge map to obtain a text vector;
specifically, inquiring and observing a patient to acquire basic information of the patient;
and reasoning the basic information according to the epilepsy knowledge graph to obtain a text vector.
An electroencephalogram image vector acquisition module 400, configured to acquire electroencephalogram image information of the patient, and to input the electroencephalogram information into a pre-constructed feature extraction model to obtain an electroencephalogram vector;
acquiring electroencephalogram information of a patient through electroencephalogram equipment and preprocessing the electroencephalogram information, wherein the electroencephalogram information comprises image information such as computed tomography, magnetic resonance imaging and electroencephalogram;
after the electroencephalogram image of a patient is acquired, preprocessing operations such as cutting and amplifying need to be carried out on the image so as to adapt to an electroencephalogram image feature extraction model.
Constructing an electroencephalogram image feature extraction model;
specifically, a feature extraction model is constructed, wherein the feature extraction model is a DCNN model;
and training the feature extraction model through a plurality of pieces of electroencephalogram image information to further obtain the electroencephalogram image feature extraction model.
And inputting the preprocessed electroencephalogram information into the electroencephalogram feature extraction model to obtain an electroencephalogram vector.
An information feature vector obtaining module 500, configured to fuse the text vector and the electroencephalogram vector to obtain an information feature vector of the patient;
a detection result obtaining module 600, configured to input the information feature vector into a pre-constructed classification model to obtain an epilepsy detection result of the patient.
Specifically, a classification model is constructed, and the classification model is trained according to a plurality of previous cases and diagnosis and treatment records, so that an epilepsy classification model is obtained;
and inputting the information characteristic vector into the epilepsy classification model to obtain the epilepsy detection result of the patient.
In this embodiment, the detection of the epilepsy is assisted by artificial intelligence, the text information and the electroencephalogram image information of the patient are fused, the type of the epilepsy is classified, the condition of the epilepsy is detected more accurately, misdiagnosis is prevented, the basis is provided for the treatment of the epilepsy of the patient, the problem of unbalanced distribution of medical resource regions is solved, the working pressure of a doctor is relieved, the diagnosis level of the doctor and the medical satisfaction degree of the patient are improved, the economic pressure of the patient is relieved, and the overall level of the epilepsy detection is improved.
Compared with the prior art, the epilepsy detecting system based on the knowledge graph and the electroencephalogram provided by the embodiment has the beneficial effects that: according to the epilepsy detecting system based on the knowledge graph and the electroencephalogram, the epilepsy is detected in an auxiliary mode by adopting a multi-mode technology, the problem that the epilepsy detecting and identifying effect is poor due to the fact that an artificial intelligence technology adopts a single-mode technology and only one characteristic can be identified, and the basic information, the physical sign information and the head electroencephalogram information of a patient cannot be well fused is avoided, specifically, the text information and the electroencephalogram information of the patient are fused, the epilepsy is diagnosed and classified according to the two information, the illness state of the patient is more accurately diagnosed, misdiagnosis is prevented, a basis is provided for treatment of the epilepsy of the patient, the intelligent level of artificial intelligence auxiliary detection is improved, the work burden of medical workers is relieved, and the technical problem that the epilepsy detecting and identifying effect is poor due to the single-mode technology is generally existed is solved.
The third embodiment of the present invention further provides a readable storage medium, on which a computer program is stored, which when executed by a processor implements the steps of the method described in the first embodiment above.
The fourth embodiment of the present invention further provides an electronic device, which includes a memory, a processor, and a computer program stored in the memory and executable on the processor, and when the processor executes the computer program, the steps of the method according to the first embodiment are implemented.
In the description of the specification, reference to the description of "one embodiment," "some embodiments," "an example," "a specific example," or "some examples" or the like means that a particular feature, structure, material, or characteristic described in connection with the embodiment or example is included in at least one embodiment or example of the invention. In this specification, the schematic representations of the terms used above do not necessarily refer to the same embodiment or example. Furthermore, the particular features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples.
The above-mentioned embodiments only express several embodiments of the present invention, and the description thereof is more specific and detailed, but not construed as limiting the scope of the present invention. It should be noted that, for those skilled in the art, various changes and modifications can be made without departing from the spirit of the invention, and these changes and modifications are all within the scope of the invention. Therefore, the protection scope of the present patent shall be subject to the appended claims.
Claims (10)
1. A epilepsy detection method based on a knowledge graph and an electroencephalogram is characterized by comprising the following steps:
acquiring epilepsy medical data, and preprocessing the epilepsy medical data;
establishing an epileptic disease knowledge map according to the preprocessed epileptic disease medical data;
acquiring basic information of a patient, and reasoning the basic information according to the epilepsy knowledge graph to acquire a text vector;
acquiring electroencephalogram information of the patient, and inputting the electroencephalogram information into a pre-constructed feature extraction model to obtain an electroencephalogram vector;
fusing the text vector and the electroencephalogram image vector to obtain an information feature vector of the patient;
and inputting the information characteristic vector into a pre-constructed classification model to obtain an epilepsy detection result of the patient.
2. The epilepsy detection method based on the knowledge-graph and the electroencephalogram according to claim 1, wherein the step of establishing the epilepsy knowledge-graph according to the preprocessed epilepsy medical data specifically comprises:
performing knowledge extraction on the preprocessed epilepsy medical data to obtain initial ternary group data, wherein the knowledge extraction comprises entity extraction, attribute extraction and relationship extraction;
after knowledge fusion is carried out on the initial ternary group data, quality evaluation is carried out on the initial ternary group data so as to obtain final ternary group data with qualified quality;
and establishing an epilepsy knowledge graph according to the final ternary group data.
3. The method for epilepsy detection based on the knowledgegraph and electroencephalogram of claim 1, wherein the step of obtaining the basic information of the patient and reasoning the basic information according to the epilepsy knowledgegraph to obtain the text vector specifically comprises:
performing inquiry observation on the patient to acquire basic information of the patient;
and reasoning the basic information according to the epilepsy knowledge graph to obtain a text vector.
4. The epilepsy detecting method based on the knowledge-graph and the electroencephalogram as claimed in claim 1, wherein the step of obtaining the electroencephalogram information of the patient and inputting the electroencephalogram information into a pre-constructed feature extraction model to obtain an electroencephalogram vector specifically comprises:
acquiring electroencephalogram information of a patient through electroencephalogram equipment and preprocessing the electroencephalogram information, wherein the electroencephalogram information comprises image information such as computed tomography, magnetic resonance imaging and electroencephalogram;
constructing an electroencephalogram image feature extraction model;
and inputting the preprocessed electroencephalogram information into the electroencephalogram feature extraction model to obtain an electroencephalogram vector.
5. The method for epilepsy detection based on knowledge-graph and electroencephalogram according to claim 1, wherein the step of inputting the information feature vector into a pre-constructed classification model to obtain the epilepsy detection result of the patient specifically comprises:
constructing a classification model, and training the classification model according to a plurality of previous cases and diagnosis and treatment records to further obtain an epilepsy classification model;
and inputting the information characteristic vector into the epilepsy classification model to obtain the epilepsy detection result of the patient.
6. The epilepsy detection method based on the knowledge-graph and the electroencephalogram according to claim 4, wherein the step of constructing the electroencephalogram feature extraction model specifically comprises:
constructing a feature extraction model, wherein the feature extraction model is a DCNN model;
and training the feature extraction model through a plurality of pieces of electroencephalogram image information to further obtain the electroencephalogram image feature extraction model.
7. The epilepsy detection method based on the knowledge-graph and the electroencephalogram according to claim 1, wherein the step of acquiring epilepsy medical data and preprocessing the epilepsy medical data specifically comprises:
acquiring epilepsy medical data, wherein the epilepsy medical data comprises medical data such as medical professional books, hospital electronic diagnosis cases, medical website knowledge and the like;
and preprocessing the epilepsy medical data to remove narrative and descriptive non-knowledge point data contents and reserve key part knowledge data.
8. A epilepsy detection system based on a knowledge-graph and an electroencephalogram is characterized in that the epilepsy detection system comprises:
the data preprocessing module is used for acquiring epilepsy medical data and preprocessing the epilepsy medical data;
the knowledge map establishing module is used for establishing an epilepsy knowledge map according to the preprocessed epilepsy medical data;
the text vector output module is used for acquiring basic information of the patient and reasoning the basic information according to the epilepsy knowledge map to obtain a text vector;
the electroencephalogram image vector acquisition module is used for acquiring electroencephalogram information of the patient and inputting the electroencephalogram information into a pre-constructed feature extraction model to obtain an electroencephalogram vector;
an information feature vector acquisition module for fusing the text vector and the electroencephalogram vector to obtain an information feature vector of the patient;
and the detection result acquisition module is used for inputting the information characteristic vector into a pre-constructed classification model so as to acquire the epilepsy detection result of the patient.
9. A readable storage medium, on which a computer program is stored, which program, when being executed by a processor, is adapted to carry out the steps of the method according to any one of claims 1 to 7.
10. An electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, the processor implementing the steps of the method according to any one of claims 1 to 7 when executing the program.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210390585.7A CN115019950A (en) | 2022-04-14 | 2022-04-14 | Epilepsia detection method, system and equipment based on knowledge graph and electroencephalogram |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210390585.7A CN115019950A (en) | 2022-04-14 | 2022-04-14 | Epilepsia detection method, system and equipment based on knowledge graph and electroencephalogram |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115019950A true CN115019950A (en) | 2022-09-06 |
Family
ID=83067406
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210390585.7A Pending CN115019950A (en) | 2022-04-14 | 2022-04-14 | Epilepsia detection method, system and equipment based on knowledge graph and electroencephalogram |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115019950A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117224080A (en) * | 2023-09-04 | 2023-12-15 | 深圳市维康致远科技有限公司 | Human body data monitoring method and device for big data |
-
2022
- 2022-04-14 CN CN202210390585.7A patent/CN115019950A/en active Pending
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117224080A (en) * | 2023-09-04 | 2023-12-15 | 深圳市维康致远科技有限公司 | Human body data monitoring method and device for big data |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111414393B (en) | Semantic similar case retrieval method and equipment based on medical knowledge graph | |
| Carchiolo et al. | Medical prescription classification: a NLP-based approach | |
| CN110556173A (en) | intelligent classification management system and method for inspection report | |
| CN116303981B (en) | Agricultural community knowledge question-answering method, device and storage medium | |
| CN112541086A (en) | Knowledge graph construction method for cerebral apoplexy | |
| CN112541066A (en) | Text-structured-based medical and technical report detection method and related equipment | |
| US20220207242A1 (en) | Method for testing medical data | |
| CN115019950A (en) | Epilepsia detection method, system and equipment based on knowledge graph and electroencephalogram | |
| CN117056493A (en) | Large language model medical question-answering system based on medical record knowledge graph | |
| Maldonado et al. | Deep learning meets biomedical ontologies: knowledge embeddings for epilepsy | |
| CN115617840A (en) | Medical data retrieval platform construction method, system, computer and storage medium | |
| CN113707304A (en) | Triage data processing method, device, equipment and storage medium | |
| Sedghi et al. | Mining clinical text for stroke prediction | |
| Baccianella et al. | Variable-constraint classification and quantification of radiology reports under the ACR Index | |
| CN111898528B (en) | Data processing method, device, computer readable medium and electronic equipment | |
| Norman | Systematic review automation methods | |
| CN116719840A (en) | Medical information pushing method based on post-medical-record structured processing | |
| CN113257371B (en) | Clinical examination result analysis method and system based on medical knowledge map | |
| Yang et al. | Process mining the trauma resuscitation patient cohorts | |
| Chou et al. | An information retrieval system for medical records & documents | |
| Sujatha et al. | Detection of Bipolar Disorder Using Machine Learning with MRI. | |
| CN113674863A (en) | Intelligent disease prevention and control and health care recommendation method for big data gene detection | |
| Chawda et al. | Learning to recommend descriptive tags for health seekers using deep learning | |
| Manivannan et al. | Health monitoring system for diabetic patients | |
| Andritsos et al. | Case-based reasoning for biomedical informatics and medicine |
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 |