CN111613339A - Similar medical record searching method and system based on deep learning - Google Patents
Similar medical record searching method and system based on deep learning Download PDFInfo
- Publication number
- CN111613339A CN111613339A CN202010412253.5A CN202010412253A CN111613339A CN 111613339 A CN111613339 A CN 111613339A CN 202010412253 A CN202010412253 A CN 202010412253A CN 111613339 A CN111613339 A CN 111613339A
- Authority
- CN
- China
- Prior art keywords
- medical record
- medical
- information
- graph
- database
- 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.)
- Granted
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
- G16H50/70—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
- 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
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
Landscapes
- Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Theoretical Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Physics & Mathematics (AREA)
- Medical Informatics (AREA)
- General Physics & Mathematics (AREA)
- Computational Linguistics (AREA)
- Biomedical Technology (AREA)
- Databases & Information Systems (AREA)
- General Health & Medical Sciences (AREA)
- Public Health (AREA)
- Life Sciences & Earth Sciences (AREA)
- General Engineering & Computer Science (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Molecular Biology (AREA)
- Animal Behavior & Ethology (AREA)
- Evolutionary Computation (AREA)
- Biophysics (AREA)
- Pathology (AREA)
- Artificial Intelligence (AREA)
- Epidemiology (AREA)
- Primary Health Care (AREA)
- Medical Treatment And Welfare Office Work (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention discloses a similar medical record searching method and system based on deep learning, which comprises the following steps: constructing a knowledge graph; extracting the subject of the medical record sample information as the characteristic information of the medical record, and storing the subject in a database; extracting medical record characteristic information from the input electronic medical record information; acquiring a sub-graph vector containing medical general knowledge related to the electronic medical record from the knowledge graph; inputting the characteristic information of the current medical record, the sub-graph vector and the characteristic information of the medical record in the medical record information sample database into a trained neural network model, and calculating the similarity between the current medical record and each medical record in the database; and outputting a set number of similar cases according to the size of the similarity. The invention has the beneficial effects that: the method comprises the steps of automatically extracting medical record features by using a siamese-transformer deep learning neural network model enhanced by a knowledge graph in the medical field, mapping medical records to the same vector space, and calculating the similarity of the two medical records by using similarity calculation in the space.
Description
Technical Field
The invention relates to the technical field of similar case finding, in particular to a similar case history finding method and system based on deep learning.
Background
The statements in this section merely provide background information related to the present disclosure and may not necessarily constitute prior art.
More and more researchers have begun to solve the problems in the medical field by using Natural Language Processing (NLP) technology. In the NLP field, text similarity is a fundamental problem, and the problem of calculating the similarity between texts still has a difficulty. Because the similarity between two sentences is measured through the semantic level, and the semantics belongs to the cognitive level, great difficulty is brought to research. Because of the current link definition, only semantic representation can be solved, and logical reasoning cannot be learned. Secondly, understanding the semantics of a sentence or a segment of a sentence does not rely on the ability to summarize the summary by induction, but also needs to be aided by external knowledge, that is, what kind of event the sentence expresses, and the event may be related to many entities, relations and paths. The relationship between two sentences, the intersection of the paths, represents the semantic contact degree to some extent.
In the prior art, the search for similar medical records is generally realized by performing matching search on one or more fields; in the medical problem, the similarity of medical texts is more difficult to calculate than in the non-specific field. This is because medical terminology is very numerous, and it is difficult to learn medical information by embedding ordinary words in vectors. This results in incomplete and inaccurate results of similar searches, which affects the efficiency of the search.
Disclosure of Invention
The invention aims to solve the defects of the prior art and provides a method and a system for searching similar medical records based on deep learning, wherein the medical record features are automatically extracted by using a deep learning model enhanced by a knowledge map in the medical field, the medical records are mapped to the same vector space, and the similarity of the two medical records is calculated by using similarity calculation in the space; similar cases can be accurately found.
In some embodiments, the following technical scheme is adopted:
a similar medical record searching method based on deep learning comprises the following steps:
constructing a knowledge graph capable of representing relationships between medical concepts;
after preprocessing the acquired medical record sample information, extracting the subject of the medical record sample information as the characteristic information of the medical record, and storing the subject in a database;
extracting medical record characteristic information from the input electronic medical record information; acquiring a sub-graph vector containing medical general knowledge related to the electronic medical record from the knowledge graph;
inputting the characteristic information of the current medical record, the sub-graph vector and the characteristic information of the medical record in the medical record information sample database into a trained neural network model, and calculating the similarity between the current medical record and each medical record in the database; and outputting a set number of similar cases according to the size of the similarity.
In other embodiments, the following technical solutions are adopted:
a similar medical record searching system based on deep learning comprises the following steps:
means for constructing a knowledge-graph capable of representing relationships between medical concepts;
the device is used for extracting the subject of the medical record sample information as the characteristic information of the medical record after preprocessing the acquired medical record sample information and storing the subject in the database;
the system is used for extracting medical record characteristic information from input electronic medical record information; means for obtaining a sub-image vector containing medical general knowledge associated with the electronic medical record in the knowledge-graph;
the system is used for inputting the characteristic information of the current medical record, the sub-graph vector and the characteristic information of the medical record in the medical record information sample database into a trained neural network model and calculating the similarity between the current medical record and each medical record in the database; and a device for outputting a set number of similar cases according to the degree of similarity.
In other embodiments, the following technical solutions are adopted:
a terminal device comprising a processor and a computer-readable storage medium, the processor being configured to implement instructions; the computer readable storage medium is used for storing a plurality of instructions, and the instructions are suitable for being loaded by a processor and executing the similar medical record searching method based on deep learning.
In other embodiments, the following technical solutions are adopted:
a computer readable storage medium, wherein a plurality of instructions are stored, and the instructions are suitable for being loaded by a processor of a terminal device and executing the similar medical record searching method based on deep learning.
Compared with the prior art, the invention has the beneficial effects that:
the method uses a siamese-transformer deep learning neural network model enhanced by a knowledge graph in the medical field to automatically extract the characteristics of medical records, maps the medical records to the same vector space, and calculates the similarity of the two medical records in the space by using similarity calculation.
The medical record query method and the medical record query system simultaneously comprise a medical record query function, similar medical records queried by a user can be retrieved, a doctor can accurately count the similar medical records, and an effective reference can be obtained for the doctor when treating similar patients. The invention improves the traditional clinical decision making process, the traditional clinical decision is judged by doctors according to own knowledge and experience, and the invention can provide effective reference when the doctors make clinical decision to improve the clinical decision of the doctors.
Drawings
FIG. 1 is a flowchart of a method for searching similar medical records of patients based on deep learning according to an embodiment of the present invention;
FIG. 2 is a diagram of a deep learning model architecture for calculating medical record similarity according to an embodiment of the present invention.
Detailed Description
It should be noted that the following detailed description is exemplary and is intended to provide further explanation of the disclosure. Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs.
It is noted that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of example embodiments according to the present application. As used herein, the singular forms "a", "an" and "the" are intended to include the plural forms as well, and it should be understood that when the terms "comprises" and/or "comprising" are used in this specification, they specify the presence of stated features, steps, operations, devices, components, and/or combinations thereof, unless the context clearly indicates otherwise.
The embodiments and features of the embodiments of the present invention may be combined with each other without conflict.
Example one
Whether the method is based on feature engineering or deep learning, the prior art rarely uses prior knowledge in research work, and the prior knowledge is very helpful for a computer to better understand the semantics of sentences, and particularly in a specific medical field, the prior knowledge graph in the professional field is used to be very helpful for calculating the similarity of medical texts.
Based on this, in one or more embodiments, a similar medical record searching method based on deep learning is disclosed, as shown in fig. 1, including the following steps:
step (1): a knowledge graph of the medical field is constructed through a knowledge graph construction technology, wherein entities in the knowledge graph represent medical concepts, and edges in the knowledge graph represent relations between the medical concepts. The constructed knowledge graph is stored in Neo4j for convenient later use.
Specifically, the method comprises the following steps:
firstly, a BERT + CRF model is trained according to some public corpus of the medical field, so that the model can identify entities and relations in the medical text. The main step is to label some data sets identified by medical concept entities manually, and in the labeling of the data sets, some common medical concepts and relationships between the medical concepts are mainly labeled. These labeled datasets are then used to train a BERT + CRF model, where BERT is collectively referred to as Bidirective encoding reproduction from transformations and CRF is collectively referred to as Conditional Random Field.
And adjusting the parameters until a model with excellent performance is obtained. The model is then used to predict the unlabeled medical text data set, and medical concepts and relationships between medical concepts in the medical text are identified.
Secondly, because the quality of the medical knowledge graph is directly related to the effect of calculating the similarity of the medical records by the model, noise data such as error entities or entities with wrong relationship are removed manually. The knowledge map finally stored in Neo4j can be more accurate and clean.
And thirdly, storing the acquired medical field knowledge graph into a Neo4j database.
Step (2): the hospital database calls the case information of the patient, and the ID, the hospitalization time and the electronic medical record of the patient and the final medical diagnosis are acquired.
Specifically, data in a hospital system is acquired, time and electronic medical record information is stored in a database, and the name of a patient needs to be represented by an automatically constructed unique ID due to desensitization of hospital data. Desensitization processing is carried out on related sensitive information, and then data after desensitization processing are stored in the constructed mysql database, so that later calling is facilitated.
And (3): and preprocessing the acquired case information, wherein the preprocessing mainly comprises removing stop words and word segmentation.
Specifically, some unusable records in some acquired electronic medical record information are deleted, because some incomplete electronic medical record information is recorded in the system in the actual operation of the hospital, the incomplete records need to be deleted, so as to prevent the information from influencing the operation of the system.
And removing stop words, presetting a dictionary of the stop words, reading information from the database, traversing the dictionary, and deleting the words same as the words in the dictionary. Some irrelevant information in some electronic medical records is removed.
And then putting the content into a word segmentation device for word segmentation. The word segmentation device uses a jieba word segmentation tool of python, and the Chinese word segmentation effect of the tool is good. And storing the preprocessed result into the mysql database, so that the system can be conveniently called later. After word segmentation, some words with wrong word segmentation are manually deleted.
And (4): the LDA technology is used for obtaining the subjects of the medical records, namely obtaining the main expression content of the medical records, so that the information which is mainly emphasized by each medical record can be probably known. The method comprises the following specific steps:
LDA is divided into training and inference, and because of the similarity of inference and training processes, the topic of the sample can be obtained for training or inferring the sample, but the training process has a great influence on the topic analysis of the whole system. For a given classification problem, it is extremely important to select an appropriate data set. This is because the subject matter analyzed from this data directly affects the learning and classification performance of the classifier. Two main conditions should be followed to construct the correct generic data set. One is that the data is large enough, and the second is that there should be a balanced distribution of words and topics (as viewed by humans) to cover the training data, and more importantly, to deal well with the diversity of future unseen data. So a sufficient number of medical record text data sets are selected as the training set of the LDA model to obtain the topic distribution of each medical record text.
Secondly, the word distribution probability of the topics is obtained through an LDA topic distribution model, wherein the best effect is found through actual calculation when the number K of the topics is 18, so that the word distribution of each topic is obtained through the following formula, then the actual meaning of each topic is manually judged according to the obtained word distribution of the topics, and each topic is manually endowed with a real meaning.
Where k is the number of topics, V is the number of words, β is a Direclet hyper-parameter (V ═ 1 … … V);
is the word tThe number of times that the theme k is assigned,
is the total number assigned to the subject K word. A two-layer for-loop needs to be designed to implement this formula. The outer layer needs to traverse each topic and the inner layer needs to traverse each vocabulary.
Thirdly, LDA obtains the theme distribution of each word in each case history text:
where zi represents the topic i, α is a Direclet hyper-parameter (K ═ 1 … … K), β is a Direclet hyper-parameter (V ═ 1 … … V);
is the number of times the word t is assigned to the topic k in addition to the current topic,
is the total number of words assigned to topic k. The formula is implemented using a for loop.
And fourthly, the LDA model obtains the theme distribution of each medical record text, and each medical record text selects a theme with the highest probability of 3 as the theme of the document.
Wherein m is the mth medical record text, K is the number of topics, K is the kth topic, and α is a Direclet hyper-parameter (K is 1 … … K); therefore, the two layers of for loops are needed, the outer layer is the number M of medical record texts, and the inner layer is the number K of subjects.
And fifthly, storing the theme of each medical record into a database as the main characteristic information of the medical record.
And (5): and calculating the similarity of medical records by using a siemese-transformer and a graphattinent model according to the case information input by the user.
Specifically, the segmented electronic medical record is used for finding a corresponding sub-graph in a constructed knowledge graph in the medical field. In order to facilitate later program invocation, the acquired subgraph is stored in the network object.
In this step, a word embedding vector is trained for each word using a larger scale unsupervised medical text data, using glove techniques from hospitals and some published medical text data sets. The trained word embedding vector can ensure that words with similar meanings are closer in the vector space.
And inputting the acquired subgraph into a graph attention model, wherein the graph attention model can acquire the characteristics of the nodes in the graph and map the subgraph into a vector, and the vector contains medical common knowledge related to the electronic medical record. Wherein the subgraph comprises nodes and edges; the nodes represent words in the electronic medical record.
Referring to fig. 2, the texts of two medical records in the training data and the theme of each medical record text are input into a siense-transformer neural network model, the siense-transformer neural network model uses a twin neural network sharing weight to calculate the similarity of the two medical records, the siense-transformer neural network model firstly uses the word embedding vector obtained in b), each word in the electronic medical records input by a user is converted into the word embedding vector through an embedding layer, the maximum number of words in the medical text is set to be 64 through analysis of the medical text, the maximum number of the words in the medical text is set, the number of the excessive parts is directly omitted, and the insufficient parts are completed by vectors initialized randomly. And then, the output of the embedded layer is input into a siemese-transformer neural network model sharing the weight, because the model has two inputs, and the transformer is the sharing weight, only one transformer network is designed to calculate the two inputs.
The obtained word embedding vector is added into a siamese-transformer network to enhance the effect of the model, the model is trained through some data with labels, and finally, the model is trained through parameter adjustment and model optimization, and the model can well check whether two medical record texts are similar or not.
A user inputs a new electronic medical record, and the electronic medical record is preprocessed, wherein the preprocessing mainly comprises word segmentation and stop word removal. Then, the vector of the electronic medical record graph is obtained, all medical records in the database are traversed, the current medical record and the medical record in the database are input into the trained deep learning model for calculating the similarity of the medical records, and the similarity of the medical record and each medical record in the database is calculated.
And (6): and displaying the first 10 similar electronic medical records from large to small according to the similarity.
In this embodiment, a query presentation interface is also designed for interaction with the user.
In the embodiment, the medical record features are automatically extracted by using a siamese-transformer deep learning model enhanced by a knowledge graph in the medical field, medical records are mapped to the same vector space, and the similarity of the two medical records is calculated by using similarity calculation in the space.
The embodiment also comprises a medical record query function, so that similar medical records queried by a user can be retrieved, doctors can accurately count the similar medical records, and effective reference can be obtained for the doctors when treating similar patients.
Example two
In one or more embodiments, a similar medical record finding system based on deep learning is disclosed, which comprises:
means for constructing a knowledge-graph capable of representing relationships between medical concepts;
the device is used for extracting the subject of the medical record sample information as the characteristic information of the medical record after preprocessing the acquired medical record sample information and storing the subject in the database;
the system is used for extracting medical record characteristic information from input electronic medical record information; means for obtaining a sub-image vector containing medical general knowledge associated with the electronic medical record in the knowledge-graph;
the system is used for inputting the characteristic information of the current medical record, the sub-graph vector and the characteristic information of the medical record in the medical record information sample database into a trained neural network model and calculating the similarity between the current medical record and each medical record in the database; and a device for outputting a set number of similar cases according to the degree of similarity.
The implementation process of the device refers to the method disclosed in the first embodiment, and details are not repeated.
EXAMPLE III
In one or more embodiments, a terminal device is disclosed, which includes a server, where the server includes a memory, a processor, and a computer program stored in the memory and executable on the processor, and the processor executes the computer program to implement the method for finding similar medical records based on deep learning in the first embodiment. For brevity, no further description is provided herein.
It should be understood that in this embodiment, the processor may be a central processing unit CPU, and the processor may also be other general purpose processors, digital signal processors DSP, application specific integrated circuits ASIC, off-the-shelf programmable gate arrays FPGA or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, and so on. A general purpose processor may be a microprocessor or the processor may be any conventional processor or the like.
The memory may include both read-only memory and random access memory, and may provide instructions and data to the processor, and a portion of the memory may also include non-volatile random access memory. For example, the memory may also store device type information.
In implementation, the steps of the above method may be performed by integrated logic circuits of hardware in a processor or instructions in the form of software.
The method for searching for similar medical records based on deep learning in the first embodiment can be directly implemented by a hardware processor, or implemented by a combination of hardware and software modules in the processor. The software modules may be located in ram, flash, rom, prom, or eprom, registers, among other storage media as is well known in the art. The storage medium is located in a memory, and a processor reads information in the memory and completes the steps of the method in combination with hardware of the processor. To avoid repetition, it is not described in detail here.
Those of ordinary skill in the art will appreciate that the various illustrative elements, i.e., algorithm steps, described in connection with the embodiments disclosed herein may be implemented as electronic hardware or combinations of computer software and electronic hardware. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the implementation. 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 application.
Although the embodiments of the present invention have been described with reference to the accompanying drawings, it is not intended to limit the scope of the present invention, and it should be understood by those skilled in the art that various modifications and variations can be made without inventive efforts by those skilled in the art based on the technical solution of the present invention.
Claims (10)
1. A similar medical record searching method based on deep learning is characterized by comprising the following steps:
constructing a knowledge graph capable of representing relationships between medical concepts;
after preprocessing the acquired medical record sample information, extracting the subject of the medical record sample information as the characteristic information of the medical record, and storing the subject in a database;
extracting medical record characteristic information from the input electronic medical record information; acquiring a sub-graph vector containing medical general knowledge related to the electronic medical record from the knowledge graph;
inputting the characteristic information of the current medical record, the sub-graph vector and the characteristic information of the medical record in the medical record information sample database into a trained neural network model, and calculating the similarity between the current medical record and each medical record in the database; and outputting a set number of similar cases according to the size of the similarity.
2. The method as claimed in claim 1, wherein a knowledge graph of the medical field is constructed, the entities in the knowledge graph represent medical concepts, and the edges in the knowledge graph represent relationships between the medical concepts.
3. The method for finding similar medical records based on deep learning of claim 1, wherein after the construction of the knowledge graph of the medical field, the method further comprises:
training a BERT + CRF model so that the model can identify entities and relationships in medical text;
predicting an unlabeled medical text data set by using the model, and identifying medical concepts in the medical text and relationships between the medical concepts;
and manually removing noise data, and storing the finally obtained knowledge graph into a database.
4. The method as claimed in claim 1, wherein the medical record information at least includes: patient name, hospital stay, electronic medical record and medical diagnosis results; a unique ID is constructed from the patient's name.
5. The method for finding similar medical records based on deep learning of claim 1, wherein the preprocessing of the acquired medical record information comprises: deleting medical record information with incomplete information, removing stop words and performing word segmentation processing.
6. The method for finding similar medical records based on deep learning as claimed in claim 1, wherein the extracting of the subject of the medical record information as the feature information of the medical record specifically comprises:
obtaining word distribution probability of the topics by using the trained text topic model, and judging the actual meaning of each topic according to the word distribution of each topic;
obtaining the topic distribution of each word in each medical record text through the text topic model;
and selecting the theme of each medical record text according to the theme distribution as the characteristic information of the medical record.
7. The method for finding similar medical records based on deep learning of claim 1, wherein the training process for the neural network model comprises:
calling feature information of two medical record samples in a database, and respectively matching corresponding sub-graphs in a knowledge graph database;
inputting the obtained sub-graph into a graph attention model, wherein the graph attention model obtains the features of the medical record theme in the graph and maps the sub-graph into a sub-graph vector, and the vector contains medical general knowledge related to the electronic medical record;
respectively inputting the characteristic information of the two medical record samples and the corresponding sub-image vectors into a neural network model, wherein the neural network model uses a twin neural network sharing weight to calculate the similarity of the information of the two medical record samples;
and repeating the process, optimizing the parameters of the neural network model, and obtaining the trained neural network model for judging whether the two medical record information are similar.
8. A similar medical record searching system based on deep learning is characterized by comprising:
means for constructing a knowledge-graph capable of representing relationships between medical concepts;
the device is used for extracting the subject of the medical record sample information as the characteristic information of the medical record after preprocessing the acquired medical record sample information and storing the subject in the database;
the system is used for extracting medical record characteristic information from input electronic medical record information; means for obtaining a sub-image vector containing medical general knowledge associated with the electronic medical record in the knowledge-graph;
the system is used for inputting the characteristic information of the current medical record, the sub-graph vector and the characteristic information of the medical record in the medical record information sample database into a trained neural network model and calculating the similarity between the current medical record and each medical record in the database; and a device for outputting a set number of similar cases according to the degree of similarity.
9. A terminal device comprising a processor and a computer-readable storage medium, the processor being configured to implement instructions; the computer-readable storage medium is used for storing a plurality of instructions, wherein the instructions are suitable for being loaded by a processor and executing the method for finding similar medical records based on deep learning according to any one of claims 1 to 7.
10. A computer-readable storage medium having stored thereon a plurality of instructions, wherein the instructions are adapted to be loaded by a processor of a terminal device and execute the method for finding similar medical records based on deep learning according to any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010412253.5A CN111613339B (en) | 2020-05-15 | 2020-05-15 | Similar medical record searching method and system based on deep learning |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010412253.5A CN111613339B (en) | 2020-05-15 | 2020-05-15 | Similar medical record searching method and system based on deep learning |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111613339A true CN111613339A (en) | 2020-09-01 |
| CN111613339B CN111613339B (en) | 2021-07-09 |
Family
ID=72201464
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010412253.5A Active CN111613339B (en) | 2020-05-15 | 2020-05-15 | Similar medical record searching method and system based on deep learning |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111613339B (en) |
Cited By (19)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112100406A (en) * | 2020-11-11 | 2020-12-18 | 腾讯科技（深圳）有限公司 | Data processing method, device, equipment and medium |
| CN112216151A (en) * | 2020-10-15 | 2021-01-12 | 北京航空航天大学 | Air traffic four-dimensional track regulation and control decision method |
| CN112542243A (en) * | 2020-12-05 | 2021-03-23 | 大连东软教育科技集团有限公司 | ICU electronic medical record knowledge graph construction method, system and storage medium |
| CN112559765A (en) * | 2020-12-11 | 2021-03-26 | 中电科大数据研究院有限公司 | Multi-source heterogeneous database semantic integration method |
| CN112650860A (en) * | 2021-01-15 | 2021-04-13 | 科技谷（厦门）信息技术有限公司 | Intelligent electronic medical record retrieval system based on knowledge graph |
| CN112687388A (en) * | 2021-01-08 | 2021-04-20 | 中山依数科技有限公司 | Interpretable intelligent medical auxiliary diagnosis system based on text retrieval |
| CN112836012A (en) * | 2021-01-25 | 2021-05-25 | 中山大学 | Similar patient retrieval method based on sequencing learning |
| CN113192626A (en) * | 2021-04-13 | 2021-07-30 | 山东大学 | Medicine taking scheme recommendation system and method based on twin neural network |
| CN113241172A (en) * | 2021-03-25 | 2021-08-10 | 边缘智能研究院南京有限公司 | ICU discrimination system for postoperative infection of neurosurgical patient |
| CN113380360A (en) * | 2021-06-07 | 2021-09-10 | 厦门大学 | Similar medical record retrieval method and system based on multi-mode medical record map |
| WO2021189971A1 (en) * | 2020-10-26 | 2021-09-30 | 平安科技（深圳）有限公司 | Medical plan recommendation system and method based on knowledge graph representation learning |
| CN113655348A (en) * | 2021-07-28 | 2021-11-16 | 国网湖南省电力有限公司 | Power equipment partial discharge fault diagnosis method based on deep twin network, system terminal and readable storage medium |
| CN113782221A (en) * | 2021-09-16 | 2021-12-10 | 平安科技（深圳）有限公司 | Disease prediction device, equipment and storage medium based on self-training learning |
| CN113782134A (en) * | 2021-09-29 | 2021-12-10 | 清华大学 | Method and system for sharing medical data |
| CN114218955A (en) * | 2021-12-28 | 2022-03-22 | 上海柯林布瑞信息技术有限公司 | Medical knowledge graph-based auxiliary reference information determination method and system |
| CN114664400A (en) * | 2022-03-18 | 2022-06-24 | 浙江星汉信息技术股份有限公司 | Medical record filing method and device |
| CN114974554A (en) * | 2022-02-23 | 2022-08-30 | 北京爱医声科技有限公司 | Method, device and storage medium for fusing atlas knowledge to strengthen medical record features |
| CN116631614A (en) * | 2023-07-24 | 2023-08-22 | 北京惠每云科技有限公司 | Treatment scheme generation method, treatment scheme generation device, electronic equipment and storage medium |
| CN116628171A (en) * | 2023-07-24 | 2023-08-22 | 北京惠每云科技有限公司 | Medical record retrieval method and system based on pre-training language model |
Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108154198A (en) * | 2018-01-25 | 2018-06-12 | 北京百度网讯科技有限公司 | Knowledge base entity normalizing method, system, terminal and computer readable storage medium |
| CN109920501A (en) * | 2019-01-24 | 2019-06-21 | 西安交通大学 | Electronic health record classification method and system based on convolutional neural networks and Active Learning |
| CN109978022A (en) * | 2019-03-08 | 2019-07-05 | 腾讯科技（深圳）有限公司 | A kind of medical treatment text message processing method and device, storage medium |
| US10360507B2 (en) * | 2016-09-22 | 2019-07-23 | nference, inc. | Systems, methods, and computer readable media for visualization of semantic information and inference of temporal signals indicating salient associations between life science entities |
| CN110265099A (en) * | 2019-05-08 | 2019-09-20 | 北京百度网讯科技有限公司 | Method and apparatus for exporting case history |
| CN110299194A (en) * | 2019-06-06 | 2019-10-01 | 昆明理工大学 | The similar case recommended method with the wide depth model of improvement is indicated based on comprehensive characteristics |
| CN110299209A (en) * | 2019-06-25 | 2019-10-01 | 北京百度网讯科技有限公司 | Similar case history lookup method, device, equipment and readable storage medium storing program for executing |
| CN110364234A (en) * | 2019-06-26 | 2019-10-22 | 浙江大学 | Electronic health record intelligent storage analyzing search system and method |
| CN110413981A (en) * | 2018-04-27 | 2019-11-05 | 阿里巴巴集团控股有限公司 | The based reminding method and device of the quality detecting method of electronic health record, similar case history |
| CN110752027A (en) * | 2019-10-21 | 2020-02-04 | 卓尔智联(武汉)研究院有限公司 | Electronic medical record data pushing method and device, computer equipment and storage medium |
| CN111105852A (en) * | 2019-12-02 | 2020-05-05 | 上海联影智能医疗科技有限公司 | Electronic medical record recommendation method and device, terminal and storage medium |
-
2020
- 2020-05-15 CN CN202010412253.5A patent/CN111613339B/en active Active
Patent Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US10360507B2 (en) * | 2016-09-22 | 2019-07-23 | nference, inc. | Systems, methods, and computer readable media for visualization of semantic information and inference of temporal signals indicating salient associations between life science entities |
| CN108154198A (en) * | 2018-01-25 | 2018-06-12 | 北京百度网讯科技有限公司 | Knowledge base entity normalizing method, system, terminal and computer readable storage medium |
| CN110413981A (en) * | 2018-04-27 | 2019-11-05 | 阿里巴巴集团控股有限公司 | The based reminding method and device of the quality detecting method of electronic health record, similar case history |
| CN109920501A (en) * | 2019-01-24 | 2019-06-21 | 西安交通大学 | Electronic health record classification method and system based on convolutional neural networks and Active Learning |
| CN109978022A (en) * | 2019-03-08 | 2019-07-05 | 腾讯科技（深圳）有限公司 | A kind of medical treatment text message processing method and device, storage medium |
| CN110265099A (en) * | 2019-05-08 | 2019-09-20 | 北京百度网讯科技有限公司 | Method and apparatus for exporting case history |
| CN110299194A (en) * | 2019-06-06 | 2019-10-01 | 昆明理工大学 | The similar case recommended method with the wide depth model of improvement is indicated based on comprehensive characteristics |
| CN110299209A (en) * | 2019-06-25 | 2019-10-01 | 北京百度网讯科技有限公司 | Similar case history lookup method, device, equipment and readable storage medium storing program for executing |
| CN110364234A (en) * | 2019-06-26 | 2019-10-22 | 浙江大学 | Electronic health record intelligent storage analyzing search system and method |
| CN110752027A (en) * | 2019-10-21 | 2020-02-04 | 卓尔智联(武汉)研究院有限公司 | Electronic medical record data pushing method and device, computer equipment and storage medium |
| CN111105852A (en) * | 2019-12-02 | 2020-05-05 | 上海联影智能医疗科技有限公司 | Electronic medical record recommendation method and device, terminal and storage medium |
Non-Patent Citations (3)
| Title |
|---|
| CHANGCHANG YIN 等: "Domain Knowledge guided deep learning with electronic health records", 《2019 IEEE INTERNATIONAL CONFERENCE ON DATA MINING (ICDM)》 * |
| FANGYUAN ZHAO 等: "Similarity Measure for Patients via A Siamese CNN Network", 《ICA3PP 2018: ALGORITHMS AND ARCHITECTURES FOR PARALLEL PROCESSING》 * |
| 董丽丽 等: "融合知识图谱与深度学习的疾病诊断方法研究", 《HTTPS://KNS.CNKI.NET/KCMS/DETAIL/11.5602.TP.20191224.1420.006.HTML》 * |
Cited By (26)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112216151A (en) * | 2020-10-15 | 2021-01-12 | 北京航空航天大学 | Air traffic four-dimensional track regulation and control decision method |
| CN112216151B (en) * | 2020-10-15 | 2021-12-28 | 北京航空航天大学 | Air traffic four-dimensional track regulation and control decision method |
| US11922817B2 (en) | 2020-10-15 | 2024-03-05 | Beihang University | 4-dimensional trajectory regulatory decision-making method for air traffic |
| WO2021189971A1 (en) * | 2020-10-26 | 2021-09-30 | 平安科技（深圳）有限公司 | Medical plan recommendation system and method based on knowledge graph representation learning |
| CN112100406A (en) * | 2020-11-11 | 2020-12-18 | 腾讯科技（深圳）有限公司 | Data processing method, device, equipment and medium |
| CN112542243B (en) * | 2020-12-05 | 2024-06-04 | 东软教育科技集团有限公司 | ICU electronic medical record knowledge graph construction method, system and storage medium |
| CN112542243A (en) * | 2020-12-05 | 2021-03-23 | 大连东软教育科技集团有限公司 | ICU electronic medical record knowledge graph construction method, system and storage medium |
| CN112559765A (en) * | 2020-12-11 | 2021-03-26 | 中电科大数据研究院有限公司 | Multi-source heterogeneous database semantic integration method |
| CN112687388A (en) * | 2021-01-08 | 2021-04-20 | 中山依数科技有限公司 | Interpretable intelligent medical auxiliary diagnosis system based on text retrieval |
| CN112687388B (en) * | 2021-01-08 | 2023-09-01 | 中山依数科技有限公司 | Explanatory intelligent medical auxiliary diagnosis system based on text retrieval |
| CN112650860A (en) * | 2021-01-15 | 2021-04-13 | 科技谷（厦门）信息技术有限公司 | Intelligent electronic medical record retrieval system based on knowledge graph |
| CN112836012A (en) * | 2021-01-25 | 2021-05-25 | 中山大学 | Similar patient retrieval method based on sequencing learning |
| CN113241172A (en) * | 2021-03-25 | 2021-08-10 | 边缘智能研究院南京有限公司 | ICU discrimination system for postoperative infection of neurosurgical patient |
| CN113192626B (en) * | 2021-04-13 | 2022-09-13 | 山东大学 | Medicine taking scheme recommendation system and method based on twin neural network |
| CN113192626A (en) * | 2021-04-13 | 2021-07-30 | 山东大学 | Medicine taking scheme recommendation system and method based on twin neural network |
| CN113380360A (en) * | 2021-06-07 | 2021-09-10 | 厦门大学 | Similar medical record retrieval method and system based on multi-mode medical record map |
| CN113655348A (en) * | 2021-07-28 | 2021-11-16 | 国网湖南省电力有限公司 | Power equipment partial discharge fault diagnosis method based on deep twin network, system terminal and readable storage medium |
| CN113655348B (en) * | 2021-07-28 | 2023-12-08 | 国网湖南省电力有限公司 | Power equipment partial discharge fault diagnosis method, system terminal and readable storage medium based on deep twin network |
| CN113782221A (en) * | 2021-09-16 | 2021-12-10 | 平安科技（深圳）有限公司 | Disease prediction device, equipment and storage medium based on self-training learning |
| CN113782134A (en) * | 2021-09-29 | 2021-12-10 | 清华大学 | Method and system for sharing medical data |
| CN114218955A (en) * | 2021-12-28 | 2022-03-22 | 上海柯林布瑞信息技术有限公司 | Medical knowledge graph-based auxiliary reference information determination method and system |
| CN114974554A (en) * | 2022-02-23 | 2022-08-30 | 北京爱医声科技有限公司 | Method, device and storage medium for fusing atlas knowledge to strengthen medical record features |
| CN114664400A (en) * | 2022-03-18 | 2022-06-24 | 浙江星汉信息技术股份有限公司 | Medical record filing method and device |
| CN116631614A (en) * | 2023-07-24 | 2023-08-22 | 北京惠每云科技有限公司 | Treatment scheme generation method, treatment scheme generation device, electronic equipment and storage medium |
| CN116628171A (en) * | 2023-07-24 | 2023-08-22 | 北京惠每云科技有限公司 | Medical record retrieval method and system based on pre-training language model |
| CN116628171B (en) * | 2023-07-24 | 2023-10-20 | 北京惠每云科技有限公司 | Medical record retrieval method and system based on pre-training language model |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111613339B (en) | 2021-07-09 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111613339B (en) | Similar medical record searching method and system based on deep learning | |
| CN111274373B (en) | Electronic medical record question-answering method and system based on knowledge graph | |
| CN110297908B (en) | Diagnosis and treatment scheme prediction method and device | |
| US10929420B2 (en) | Structured report data from a medical text report | |
| KR102153920B1 (en) | System and method for interpreting medical images through the generation of refined artificial intelligence reinforcement learning data | |
| CN112002411A (en) | Cardiovascular and cerebrovascular disease knowledge map question-answering method based on electronic medical record | |
| CN110675944A (en) | Triage method and device, computer equipment and medium | |
| CN109871538A (en) | A kind of Chinese electronic health record name entity recognition method | |
| CN113241135A (en) | Disease risk prediction method and system based on multi-mode fusion | |
| CN111402979B (en) | Method and device for detecting consistency of disease description and diagnosis | |
| JP2020516997A (en) | System and method for model-assisted cohort selection | |
| CN111274806A (en) | Method and device for recognizing word segmentation and part of speech and method and device for analyzing electronic medical record | |
| CN111368094A (en) | Entity knowledge map establishing method, attribute information acquiring method, outpatient triage method and device | |
| WO2023029502A1 (en) | Method and apparatus for constructing user portrait on the basis of inquiry session, device, and medium | |
| CN110866124A (en) | Medical knowledge graph fusion method and device based on multiple data sources | |
| CN112151183A (en) | Entity identification method of Chinese electronic medical record based on Lattice LSTM model | |
| CN113764112A (en) | Online medical question and answer method | |
| CN112347781A (en) | Generating or modifying ontologies representing relationships within input data | |
| CN115841861A (en) | Similar medical record recommendation method and system | |
| CN118114675B (en) | Medical named entity recognition method and device based on large language model | |
| CN116992002A (en) | Intelligent care scheme response method and system | |
| CN117854715B (en) | Intelligent diagnosis assisting system based on inquiry analysis | |
| CN117594206A (en) | Patient integrated triage system and method based on medical interconnection platform | |
| CN118016224A (en) | Intelligent case analysis auxiliary system based on medical big data platform | |
| CN114298314A (en) | Multi-granularity causal relationship reasoning method based on electronic medical record |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |