CN112117009A - Method, device, electronic equipment and medium for constructing label prediction model - Google Patents
Method, device, electronic equipment and medium for constructing label prediction model Download PDFInfo
- Publication number
- CN112117009A CN112117009A CN202011022926.2A CN202011022926A CN112117009A CN 112117009 A CN112117009 A CN 112117009A CN 202011022926 A CN202011022926 A CN 202011022926A CN 112117009 A CN112117009 A CN 112117009A
- Authority
- CN
- China
- Prior art keywords
- disease
- label
- medical record
- prediction model
- actual
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 54
- 201000010099 disease Diseases 0.000 claims abstract description 224
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 claims abstract description 224
- 238000003745 diagnosis Methods 0.000 claims abstract description 21
- 238000010276 construction Methods 0.000 claims abstract description 12
- 230000015654 memory Effects 0.000 claims description 20
- 238000012545 processing Methods 0.000 claims description 20
- 230000009467 reduction Effects 0.000 claims description 15
- 230000002159 abnormal effect Effects 0.000 claims description 8
- 238000013528 artificial neural network Methods 0.000 claims description 6
- 208000026350 Inborn Genetic disease Diseases 0.000 claims description 4
- 208000016361 genetic disease Diseases 0.000 claims description 4
- 230000005856 abnormality Effects 0.000 claims description 2
- 238000003058 natural language processing Methods 0.000 abstract description 5
- 238000013473 artificial intelligence Methods 0.000 abstract description 2
- 238000013527 convolutional neural network Methods 0.000 description 9
- 230000008569 process Effects 0.000 description 9
- 238000004891 communication Methods 0.000 description 7
- 238000012549 training Methods 0.000 description 7
- 230000006870 function Effects 0.000 description 6
- 238000011176 pooling Methods 0.000 description 6
- 238000004590 computer program Methods 0.000 description 4
- 230000007547 defect Effects 0.000 description 4
- 230000000694 effects Effects 0.000 description 4
- 206010067197 Tinea manuum Diseases 0.000 description 3
- 238000012937 correction Methods 0.000 description 3
- 238000010586 diagram Methods 0.000 description 3
- 208000002173 dizziness Diseases 0.000 description 3
- 238000000605 extraction Methods 0.000 description 3
- 239000013598 vector Substances 0.000 description 3
- 206010011224 Cough Diseases 0.000 description 2
- 206010024453 Ligament sprain Diseases 0.000 description 2
- 201000002481 Myositis Diseases 0.000 description 2
- 230000001154 acute effect Effects 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 2
- 230000002490 cerebral effect Effects 0.000 description 2
- 206010008118 cerebral infarction Diseases 0.000 description 2
- 208000026106 cerebrovascular disease Diseases 0.000 description 2
- 230000019771 cognition Effects 0.000 description 2
- 239000003814 drug Substances 0.000 description 2
- 229940079593 drug Drugs 0.000 description 2
- 230000002183 duodenal effect Effects 0.000 description 2
- 210000001198 duodenum Anatomy 0.000 description 2
- 208000026278 immune system disease Diseases 0.000 description 2
- 230000003993 interaction Effects 0.000 description 2
- 239000004973 liquid crystal related substance Substances 0.000 description 2
- 238000005065 mining Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000002360 preparation method Methods 0.000 description 2
- 210000002784 stomach Anatomy 0.000 description 2
- 208000024891 symptom Diseases 0.000 description 2
- 201000004647 tinea pedis Diseases 0.000 description 2
- 238000011282 treatment Methods 0.000 description 2
- 206010061424 Anal cancer Diseases 0.000 description 1
- 206010002153 Anal fissure Diseases 0.000 description 1
- 206010059313 Anogenital warts Diseases 0.000 description 1
- 208000003343 Antiphospholipid Syndrome Diseases 0.000 description 1
- 208000007860 Anus Neoplasms Diseases 0.000 description 1
- 208000016583 Anus disease Diseases 0.000 description 1
- 208000004020 Brain Abscess Diseases 0.000 description 1
- 206010007281 Carcinoid tumour of the stomach Diseases 0.000 description 1
- 208000032170 Congenital Abnormalities Diseases 0.000 description 1
- 208000002330 Congenital Heart Defects Diseases 0.000 description 1
- 206010010356 Congenital anomaly Diseases 0.000 description 1
- 208000011231 Crohn disease Diseases 0.000 description 1
- 206010011732 Cyst Diseases 0.000 description 1
- 208000009531 Fissure in Ano Diseases 0.000 description 1
- 206010061968 Gastric neoplasm Diseases 0.000 description 1
- 208000032612 Glial tumor Diseases 0.000 description 1
- 206010018338 Glioma Diseases 0.000 description 1
- 208000013875 Heart injury Diseases 0.000 description 1
- 201000006219 Herpangina Diseases 0.000 description 1
- 208000029462 Immunodeficiency disease Diseases 0.000 description 1
- 208000005168 Intussusception Diseases 0.000 description 1
- 206010023856 Laryngeal squamous cell carcinoma Diseases 0.000 description 1
- 208000005228 Pericardial Effusion Diseases 0.000 description 1
- 206010056658 Pseudocyst Diseases 0.000 description 1
- 208000010040 Sprains and Strains Diseases 0.000 description 1
- 208000000102 Squamous Cell Carcinoma of Head and Neck Diseases 0.000 description 1
- 206010002168 anal polyp Diseases 0.000 description 1
- 210000000436 anus Anatomy 0.000 description 1
- 201000011165 anus cancer Diseases 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 238000013475 authorization Methods 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 210000004204 blood vessel Anatomy 0.000 description 1
- 210000000988 bone and bone Anatomy 0.000 description 1
- 210000004556 brain Anatomy 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 201000011510 cancer Diseases 0.000 description 1
- 230000001364 causal effect Effects 0.000 description 1
- 230000001413 cellular effect Effects 0.000 description 1
- 238000012512 characterization method Methods 0.000 description 1
- 239000003795 chemical substances by application Substances 0.000 description 1
- 208000028831 congenital heart disease Diseases 0.000 description 1
- 208000031513 cyst Diseases 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 210000003195 fascia Anatomy 0.000 description 1
- 239000000835 fiber Substances 0.000 description 1
- 210000002683 foot Anatomy 0.000 description 1
- 230000002496 gastric effect Effects 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 208000015181 infectious disease Diseases 0.000 description 1
- 206010022000 influenza Diseases 0.000 description 1
- 230000007774 longterm Effects 0.000 description 1
- 206010025135 lupus erythematosus Diseases 0.000 description 1
- 210000002751 lymph Anatomy 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 230000036244 malformation Effects 0.000 description 1
- 238000010295 mobile communication Methods 0.000 description 1
- 210000003205 muscle Anatomy 0.000 description 1
- 210000005036 nerve Anatomy 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000012805 post-processing Methods 0.000 description 1
- 210000004203 pyloric antrum Anatomy 0.000 description 1
- 238000011946 reduction process Methods 0.000 description 1
- 210000002345 respiratory system Anatomy 0.000 description 1
- 230000001953 sensory effect Effects 0.000 description 1
- 210000003491 skin Anatomy 0.000 description 1
- 239000007787 solid Substances 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
- 239000002699 waste material Substances 0.000 description 1
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
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
Landscapes
- Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Health & Medical Sciences (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Biomedical Technology (AREA)
- General Health & Medical Sciences (AREA)
- Computing Systems (AREA)
- Software Systems (AREA)
- Evolutionary Computation (AREA)
- Biophysics (AREA)
- Molecular Biology (AREA)
- Artificial Intelligence (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Mathematical Physics (AREA)
- Computational Linguistics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Public Health (AREA)
- Medical Informatics (AREA)
- Pathology (AREA)
- Databases & Information Systems (AREA)
- Epidemiology (AREA)
- Primary Health Care (AREA)
- Medical Treatment And Welfare Office Work (AREA)
Abstract
The embodiment of the application discloses a method, a device, electronic equipment and a computer-readable storage medium for constructing a label prediction model, relates to the field of artificial intelligence, and particularly relates to natural language processing, a knowledge graph and big data, and can be applied to an intelligent medical scene. One embodiment of the method comprises: establishing a disease and category label system according to a first sample recorded with the corresponding relation between specific diseases and disease categories; establishing a disease and part label system according to a second sample recorded with the corresponding relation between the specific disease and the human body part; establishing a medical record and disease label system according to a third sample recorded with the corresponding relation between the historical medical record and the historical diagnosis disease; and finally, constructing a label prediction model according to the disease and category label system, the disease and part label system and the medical record and disease label system. The implementation mode provides a construction method of a multi-task learning prediction model, and the output prediction label can be more comprehensive and accurate.
Description
Technical Field
The application relates to the field of artificial intelligence, in particular to natural language processing, a knowledge graph and big data, which can be applied to the field of intelligent medical scenes, and particularly relates to a method, a device, electronic equipment and a computer-readable storage medium for constructing a label prediction model.
Background
With the advance of electronic informatization towards various industries, more and more quantities are gathered on the line, and various models can be constructed by effectively utilizing a large amount of user data by combining the concept of machine learning so as to realize various purposes including data prediction and content analysis.
In the prior art, a mode of dispersing different purposes into a plurality of single prediction models is generally adopted, that is, different purpose parameters are respectively predicted through different single prediction models.
Disclosure of Invention
The embodiment of the application provides a method and a device for constructing a label prediction model, electronic equipment and a computer-readable storage medium.
In a first aspect, an embodiment of the present application provides a method for building a label prediction model, including: establishing a disease and category label system according to a first sample recorded with the corresponding relation between specific diseases and disease categories; establishing a disease and part label system according to a second sample recorded with the corresponding relation between the specific disease and the human body part; establishing a medical record and disease label system according to a third sample recorded with the corresponding relation between the historical medical record and the historical diagnosis disease; and constructing a label prediction model according to the disease and category label system, the disease and part label system and the medical record and disease label system.
In a second aspect, an embodiment of the present application provides an apparatus for building a tag prediction model, including: a disease and category label system establishing unit configured to establish a disease and category label system according to a first sample in which a correspondence relationship between a specific disease and a disease category is recorded; a disease and part label system establishing unit configured to establish a disease and part label system based on the second sample in which the correspondence between the specific disease and the part of the human body is recorded; a medical record and disease label system establishing unit configured to establish a medical record and disease label system according to a third sample recorded with a corresponding relationship between a historical medical record and a historical diagnosis disease; a label prediction model construction unit configured to construct a label prediction model according to the disease and category label system, the disease and location label system, and the medical record and disease label system.
In a third aspect, an embodiment of the present application provides an electronic device, including: at least one processor; and a memory communicatively coupled to the at least one processor; wherein the memory stores instructions executable by the at least one processor to cause the at least one processor to perform the method for building a label prediction model as described in any one of the implementations of the first aspect when executed.
In a fourth aspect, embodiments of the present application provide a non-transitory computer-readable storage medium storing computer instructions for enabling a computer to implement a method for constructing a tag prediction model as described in any implementation manner of the first aspect when executed.
According to the method, the device, the electronic equipment and the computer-readable storage medium for constructing the label prediction model, firstly, a disease and category label system is established according to a first sample recorded with the corresponding relation between specific diseases and disease categories; then, establishing a disease and part label system according to a second sample recorded with the corresponding relation between the specific disease and the human body part; then, establishing a medical record and disease label system according to a third sample recorded with the corresponding relation between the historical medical record and the historical diagnosis disease; and finally, constructing a label prediction model according to the disease and category label system, the disease and part label system and the medical record and disease label system.
According to the technical scheme, the construction scheme of the multi-task learning prediction model is provided, namely the first sample, the second sample and the third sample are used as training samples at the same time, the latent commonalities and incidence relations among different samples can be learned at the same time under different task targets, the constructed multi-task learning label prediction model can output more comprehensive and accurate prediction labels, and therefore the technical defects that data are not shared and the accuracy of prediction results is not high in the prior art that a plurality of single-task learning models are obtained through respective construction and then are summarized are overcome.
It should be understood that the statements in this section do not necessarily identify key or critical features of the embodiments of the present disclosure, nor do they limit the scope of the present disclosure. Other features of the present disclosure will become apparent from the following description.
Drawings
Other features, objects and advantages of the present application will become more apparent upon reading of the following detailed description of non-limiting embodiments thereof, made with reference to the accompanying drawings in which:
FIG. 1 is an exemplary system architecture to which the present application may be applied;
fig. 2 is a flowchart of a method for constructing a tag prediction model according to an embodiment of the present application;
FIG. 3 is a flow chart of another method for constructing a label prediction model provided by an embodiment of the present application;
FIG. 4 is a schematic flowchart of a method for constructing a label prediction model in an application scenario according to an embodiment of the present application;
fig. 5 is a block diagram illustrating an apparatus for building a tag prediction model according to an embodiment of the present application;
fig. 6 is a schematic structural diagram of an electronic device suitable for executing a method for building a tag prediction model according to an embodiment of the present application.
Detailed Description
The present application will be described in further detail with reference to the following drawings and examples. It is to be understood that the specific embodiments described herein are merely illustrative of the relevant invention and not restrictive of the invention. It should be noted that, for convenience of description, only the portions related to the related invention are shown in the drawings.
It should be noted that the embodiments and features of the embodiments in the present application may be combined with each other without conflict. The present application will be described in detail below with reference to the embodiments with reference to the attached drawings.
Fig. 1 illustrates an exemplary system architecture 100 to which embodiments of the methods, apparatuses, electronic devices and computer-readable storage media for building a label prediction model of the present application may be applied.
As shown in fig. 1, the system architecture 100 may include terminal devices 101, 102, 103, a network 104, and a server 105. The network 104 serves as a medium for providing communication links between the terminal devices 101, 102, 103 and the server 105. Network 104 may include various connection types, such as wired, wireless communication links, or fiber optic cables, to name a few.
The user may use the terminal devices 101, 102, 103 to interact with the server 105 via the network 104 to receive or send messages or the like. The terminal devices 101, 102, 103 and the server 105 may be installed with various applications for implementing information communication between the two devices, such as a data transmission application, a prediction model building application, an instant messaging application, and the like.
The terminal apparatuses 101, 102, 103 and the server 105 may be hardware or software. When the terminal devices 101, 102, 103 are hardware, they may be various electronic devices with display screens, including but not limited to smart phones, tablet computers, laptop portable computers, desktop computers, and the like; when the terminal devices 101, 102, and 103 are software, they may be installed in the electronic devices listed above, and they may be implemented as multiple software or software modules, or may be implemented as a single software or software module, and are not limited in this respect. When the server 105 is hardware, it may be implemented as a distributed server cluster composed of multiple servers, or may be implemented as a single server; when the server is software, the server may be implemented as a plurality of software or software modules, or may be implemented as a single software or software module, which is not limited herein.
The server 105 may provide various services through various built-in applications, taking as an example that a prediction model building class application for building a label prediction model service may be provided, the server 105 may implement the following effects when running the class application: first, a first sample in which a correspondence between a specific disease and a disease category is recorded, a second sample in which a correspondence between a specific disease and a human body part is recorded, and a third sample in which a correspondence between a history medical record and a history diagnosis disease is recorded are acquired from terminal devices 101, 102, 103 through a network 104; then, respectively establishing a disease and category label system, a disease and part label system and a medical record and disease label system according to the first sample, the second sample and the third sample; and finally, constructing a label prediction model according to the disease and category label system, the disease and part label system and the medical record and disease label system. Further, after the server 105 constructs an available label prediction model, a label prediction service based on the label prediction model may be provided to the outside, for example, an actual electronic medical record sent from the terminal device 101, 102, 103 through the network 104 by the user is received, the actual electronic medical record is input into the label prediction model, and the output actual disease category label and the actual body part label are returned to the user.
It should be noted that the first sample in which the correspondence between the specific disease and the disease type is recorded, the second sample in which the correspondence between the specific disease and the human body part is recorded, and the third sample in which the correspondence between the historical medical record and the historical diagnosis disease is recorded may be acquired from the terminal apparatuses 101, 102, and 103 through the network 104, or may be stored locally in the server 105 in advance in various ways. Thus, when the server 105 detects that such data is already stored locally (e.g., a backup is left in the server's local storage unit), the data may be selected to be retrieved directly from the local, in which case the exemplary system architecture 100 may also not include the terminal devices 101, 102, 103 and the network 104.
Since a large number of corresponding samples are required to be used for training to obtain a label prediction model capable of outputting a plurality of types of labels through learning and training, and a large number of computing resources of equipment are required to be occupied and a strong computing capability is required, the method for constructing the label prediction model provided in the subsequent embodiments of the present application is generally executed by the server 105 having a strong computing capability and a large number of computing resources, and accordingly, the device for constructing the label prediction model is generally arranged in the server 105.
It should be understood that the number of terminal devices, networks, and servers in fig. 1 is merely illustrative. There may be any number of terminal devices, networks, and servers, as desired for implementation.
Referring to fig. 2, fig. 2 is a flowchart of a method for constructing a tag prediction model according to an embodiment of the present application, wherein the process 200 includes the following steps:
step 201: establishing a disease and category label system according to a first sample recorded with the corresponding relation between specific diseases and disease categories;
this step is intended to establish a disease and category label system from a first sample in which correspondence between a specific disease and a disease category is recorded by an executing agent (for example, the server 105 shown in fig. 1) of the method for constructing a label prediction model.
The first sample recorded with the correspondence between the specific disease and the disease category may be obtained through various channels, such as extraction from a medical book, acquisition from data annotation of an experienced physician, acquisition from a knowledge graph related to the medical disease issued by the lawyer, and the like, and an appropriate acquisition mode may be selected according to actual situations, which is not specifically limited herein. Further, according to different sources of the first sample, the execution subject should also select an acquisition manner corresponding to the first sample, for example, for a resource disclosed on the network, the execution subject can directly acquire the resource from the network through a web crawler or a downloading manner, and for a resource stored on-line, the execution subject can legally acquire the resource through an authorization application manner.
It should be understood that a specific disease may belong to different disease categories at the same time, and each disease category may include a plurality of different specific diseases, so that the correspondence between the specific disease and the disease category described in this step is actually a complex correspondence including one-to-one, one-to-many, many-to-one, and many-to-many, and therefore, a recording manner capable of recording many-to-many correspondence in a simplified manner should be selected as much as possible in selecting the recording manner.
A data set for extracting the first sample, which records the correspondence between specific diseases and disease categories in a tabular form, is given below, see table 1 below:
TABLE 1 Table of correspondence between specific diseases and disease categories
Table 1 above records the inclusion relationship between the partial disease category and the included partial specific disease in a table format, taking the disease category of "immune disease" as an example, table 1 specifically includes four specific diseases of "myositis", "lupus erythematosus", "antiphospholipid syndrome" and "immunodeficiency disease", i.e., shows a one-to-many correspondence relationship, and the constructed disease and category label system may be a system in which the above contents are labeled and expressed, so that the correspondence relationship between the category label and the specific disease label is used to establish the system.
Specifically, "immune diseases" can be directly used as a category label of the category diseases, and "myositis" can be used as a label of a specific disease, or for the sake of brevity, the labels are respectively represented as short and unique specific character strings or numbers, so that the storage and retrieval performance of the corresponding relationship can be increased while the same effect is achieved. In addition, the tag system is constructed by the characteristic that the encapsulated content representation form of the tag is convenient to call and perform post-processing, so that the efficiency is improved.
Step 202: establishing a disease and part label system according to a second sample recorded with the corresponding relation between the specific disease and the human body part;
in this step, the executing entity establishes a disease and category label system according to the first sample recorded with the corresponding relationship between the specific disease and the disease category. Except that the corresponding relationship among the records in the obtained sample is different, the rest is referred to the related description in step 201.
Data that can be used to extract a second sample, which records the correspondence between specific diseases and body parts in tabular form, is given below, see table 2 below:
TABLE 2 Table of correspondence between specific diseases and body parts
Parts of human body | Specific diseases |
Heart and heart | Congenital heart disease of newborn …, heart injury, pericardial effusion |
Duodenum | Duodenal bulb, duodenum intussusception, duodenal Crohn's disease … |
Ear part | Congenital auricular deformity, otogenic brain abscess, pseudocyst of auricle … |
Hand part | … for tinea manuum, tinea pedis, sprain and strain of finger |
Stomach (stomach) | … for gastric carcinoid, gastric tumor, gastric antrum and gastric body malignant tumor |
Brain part | … for cerebellar cyst, cerebral infarction, cerebral glioma, and cerebral infarction |
Anus | Anal polyp, anogenital condyloma acuminata, anal cancer, anal fissure … |
Throat area | Herpangina, squamous cell carcinoma of larynx, pharyngeal malformation … |
Table 2 above records the inclusion relationship between a part of a human body and a part of a specific disease included in the part of the human body in a table form, taking a human body part such as "ear" as an example, table 2 specifically includes four specific diseases such as "tinea manus", "tinea manus and pedis", "finger sprain" and "strain", i.e. the four specific diseases are expressed as a one-to-many correspondence relationship, and the constructed disease and part label system may be a system in which the above contents are labeled and thus the correspondence relationship between the part label and the specific disease label is used to establish the system.
Further, the human body part can be subdivided into multiple stages, for example, the whole body part of the human body can be divided into several parts, such as skin, fascia, muscle, blood vessel, lymph, nerve, bone and joint, or divided according to a two-stage part system shown in the following table 3:
TABLE 3 Secondary division system for human body parts
Accordingly, it can be seen that the correspondence between the parts and the specific diseases shown in table 2 is only a result of one of the part division modes, and different forms can be obtained according to the division of the human body parts in the actual application scene, which is not specifically limited herein.
Step 203: establishing a medical record and disease label system according to a third sample recorded with the corresponding relation between the historical medical record and the historical diagnosis disease;
different from the two label systems established in step 201 and step 202, the step aims to establish a medical record and disease label system by the executing subject according to a third sample recorded with the corresponding relationship between the historical medical record and the historical diagnosis disease, so as to find the association between the medical record content described by the natural language and the actual disease diagnosis result through a large amount of sample data by a machine and a model.
Specifically, no matter paper medical records or electronic medical records, a content writer usually expresses based on self cognition on own body states, and considering that different users have different cognitions, in order to fully mine potential associations between history medical records and diagnosis diseases as much as possible and find commonalities from special cases based on the potential associations, medical record contents can be processed in various ways, including but not limited to medical entity extraction, symptom element extraction, character correction, confidence determination and the like, and because characters are inconvenient to perform feature operations, texts can be converted into other expression forms to participate in operations such as feature matching and the like, such as vectors, high-dimensional features and the like, in a proper way.
It should be noted that, in this embodiment, there is no causal relationship among the steps 201, 202, and 203, and all the steps may be executed separately or in parallel, and this embodiment only provides an execution mode that appears to be executed in series for convenience of a flow-based execution mode.
Step 204: and constructing a label prediction model according to the disease and category label system, the disease and part label system and the medical record and disease label system.
In step 201, step 202 and step 203, the execution subject constructs a label prediction model according to a disease and category label system, a disease and part label system and a medical record and disease label system.
In other words, in this step, the correspondence between specific diseases and disease categories, the correspondence between specific diseases and human body parts, and the correspondence between historical medical records and historical diagnosis diseases are used as training data to construct a multi-task learning model, i.e., the label prediction model. The reason why the label prediction model is obtained is that the prediction model is constructed based on the label system and outputs the label with the corresponding content.
The method for constructing the label prediction model provided by the embodiment of the application actually provides a construction scheme of the multi-task learning prediction model, namely, the first sample, the second sample and the third sample are simultaneously used as training samples, so that the hidden commonalities and incidence relations among different samples can be learned at the same time under different task targets, the constructed multi-task learning label prediction model can be more comprehensive and accurate based on the prediction labels output by actual medical records, and the technical defects that data are not shared and the accuracy of prediction results is not high in the process of respectively constructing and obtaining a plurality of single-task learning models in the prior art can be improved as far as possible.
Referring to fig. 3, fig. 3 is a flowchart of another method for constructing a tag prediction model according to an embodiment of the present application, wherein the process 300 includes the following steps:
step 301: establishing a disease and category label system according to a first sample recorded with the corresponding relation between specific diseases and disease categories;
step 302: establishing a disease and part label system according to a second sample recorded with the corresponding relation between the specific disease and the human body part;
step 303: establishing a medical record and disease label system according to a third sample recorded with the corresponding relation between the historical medical record and the historical diagnosis disease;
in some other embodiments of the present application, a method for specifically implementing establishment of a medical record and disease label system is further provided, and a main focus is on how to extract and process information from a historical medical record, which specifically includes the following steps:
1) extracting natural language texts from historical medical records, and determining suspected diseases according to historical diagnosis diseases;
2) performing feature dimension reduction processing on the natural language text to obtain dimension reduction text features;
the purpose of the dimension reduction processing is to convert original high-dimensional features into low-dimensional features, so that the difficulty of mining the contents of the features is reduced, and some complex underlying features are exposed. In particular, the dimension reduction process used can be implemented using an Embedding layer (transliteration into an Embedding layer).
3) Processing the dimensionality reduction text features by using a preset neural network to obtain processed medical record features;
reduced-dimension text features can be processed, for example, by Convolutional Neural Networks (CNN), which is a type of feed-forward Neural network that contains Convolutional calculations and has a deep structure and is one of the representative algorithms for deep learning. CNNs are generally composed of convolutional layers, pooling layers, and fully-connected layers, and convolutional neural networks have a characterization learning capability and can perform translation-invariant classification on input information according to their hierarchical structures. This step aims to extract more features from the CNN data that can be used to find commonalities with suspected diseases by convolution, pooling of the data.
4) And establishing a label system for obtaining the medical record and the disease according to the labeled representation of the suspected disease corresponding to the processed medical record characteristics.
Tagging may be understood as a way of encapsulating content to facilitate invocation and matching in the manner of a tag.
The data processing mode for sequentially performing the dimensionality reduction processing and the neural network processing on the natural language texts contained in the medical record contents can be used for mining the medical record contents as deeply as possible, so that the relevance between the medical record contents and the suspected diseases can be found as comprehensively and accurately as possible.
Step 304: constructing a label prediction model according to a disease and category label system, a disease and part label system and a medical record and disease label system;
the steps 301-304 are the same as the steps 201-204 shown in fig. 2, and the contents of the same portions refer to the corresponding portions of the previous embodiment, which are not described herein again.
The above sections all belong to the content of how to construct and obtain the label prediction model, and the following steps will be described with respect to how to actually use the constructed label prediction model, it should be understood that the use manner given in the following steps does not need to repeat the above construction steps of the label prediction model before each use unless a new application scenario unsuitable for the previously constructed label prediction model is replaced.
Step 305: receiving an incoming actual electronic medical record;
step 306: determining an actual disease category label and an actual human body part label corresponding to an actual electronic medical record by using a label prediction model;
the step aims to determine an actual disease category label and an actual human body part label corresponding to an actual electronic medical record by the executing body by utilizing a label prediction model which is constructed in advance. In a specific implementation level, the actual electronic medical record can be imported into the label prediction model as input data, and the actual disease category label and the actual human body part label output by the label prediction model are received.
It should be understood that the reason why the label prediction model can output the actual disease category label and the actual human body part label is that, in the construction process of the label prediction model, the capability of determining the suspected disease corresponding to the medical record content is learned from the medical record and diagnosis disease label system, and on the basis of knowing the suspected disease, the corresponding actual disease category is determined by the membership relationship between the specific disease and disease category learned in the disease and part label system, and the corresponding human body part is determined by the correspondence relationship between the specific disease and human body part learned in the disease and part label system. Furthermore, the determined label representing the corresponding disease category and the label of the human body part can also be used for reversely demonstrating the accuracy of suspected diseases.
Step 307: determining a corresponding user attribute label according to a natural language text of an actual electronic medical record;
the step is to determine corresponding user attribute labels by the executing body according to the natural language text of the actual electronic medical record, wherein the user attribute labels comprise sex labels, family genetic disease labels, working environment labels and the like for representing personal attributes of the user, and the attribute labels are used for participating in other data processing as reference influence factors, such as data statistics, whether family diseases exist, whether heredity exists and the like.
Step 308: and returning an actual disease category label, an actual human body part label and a user attribute label as an attached label according to a preset path.
On the basis of step 306 and step 307, this step is intended to return the actual disease category label, the actual body part label, and the user attribute label as the attached label by a preset path. The preset path includes, but is not limited to, a mail, a short message, an interface popup, an instant messaging application, and the like.
On the basis of the previous embodiment, the embodiment not only provides a specific implementation mode for establishing a label system of medical records and diagnosing diseases, but also can more comprehensively find the relevance between the medical records and suspected diseases by carrying out various treatments on the medical record contents; meanwhile, a use mode of how to output an actual label corresponding to the actual electronic medical record by using the constructed label prediction model is given through the steps 305 to 308.
It should be noted that the technical solution of outputting the actual disease category label and the actual human body part label through the label prediction model given in step 306 and the technical solution of analyzing the user attribute label extracted from the natural language text extracted from the medical record content given in step 307 can completely and separately form two different label return solutions, that is, performing two analyses and returning two types of labels simultaneously as in this embodiment, or in another embodiment, only including the technical solution as in step 306, and subsequently returning only the actual disease category label and the actual human body part label output by the label prediction model. The purpose of adding step 307 is to enable more convenient and comprehensive performance of some other statistics based on a one's attached label.
In some other embodiments of the present application, one statistical approach to labels that includes and is not limited to may be:
counting quantity abnormal labels of which the actual output quantity exceeds the preset quantity in a preset time period;
and determining whether the outbreak phenomenon of the short-term epidemic exists according to the quantity abnormal label.
For example, in a month, if the disease category label of the upper respiratory tract acute infection exists in more than 80% of labels output by electronic medical records, and the average is far more than 40%, the phenomenon of acute influenza outbreak can be considered to exist to some extent. Further, a targeted error may be made for such a situation found, such as a push precaution.
It should be understood that the output label can be used for multiple purposes, such as counting to estimate the number of patients and the conditions of the patients in a certain area, arranging the drug preparation in a pharmacy or drug preparation institution, reducing waste, etc.
On the basis of any embodiment, in order to guarantee the long-term usability of the label prediction model, the user can be allowed to make correction and error identification on the real situation of the output prediction label, and the parameters of the model can be timely adjusted according to the correction information and the error identification, so that the model which is continuously corrected has better prediction accuracy. One specific implementation may be: and receiving label error information transmitted by the label with the prediction error, and adjusting the parameters of the label prediction model according to the label error information.
For further understanding, the present application also provides a specific implementation scheme in combination with a specific application scenario, please refer to a data processing flow chart for implementing the tag prediction effect as shown in fig. 4:
the meaning of english appearing in fig. 4 is explained first:
NLU is English abbreviation of Natural Language Understanding, Chinese is translated into Natural Language Understanding, and aims to explain that the input content of the medical record is correspondingly processed; text herein refers to natural text of medical records obtained after natural language processing, such as cough 3 days, dizziness 2 days; feat refers to medical record characteristics (such as symptoms, signs, diseases, etc.) extracted from the results after natural language processing, and corresponds to the above two feats, i.e. "cough" and "dizziness"; the flat-part refers to the part information to which flat belongs, for example, the part information of dizziness is the head; sex refers to the sex information of the patient; popul refers to the demographic attribute information of the patient, e.g., adult, child, pregnant; the emb refers to text distributed semantic coding Embedding processing and is a processing mode for performing dimension reduction processing on features; average pooling treatment of avg pooling value; CNN refers to convolutional neural networks, the left part of FIG. 4 shows a typical CNN network hierarchy, where filter refers to the filter, Conv refers to the convolutional layer, relu is a loss function, and max-pooling refers to the max pooling operation; fc denotes the full link layer; one-hot refers to one-hot encoding of text features, such as gender one-hot encoding: the male one-hot code is (1, 0), and the female one-hot code is (0, 1); "+" means that a plurality of features expressed in vector form are spliced; sigmoid refers to the loss function used by the model.
The whole using process comprises the following steps:
1) the server receives an incoming natural text of the medical record;
2) the server processes the natural text of the medical record through a natural language processing technology and analyzes the natural text (text) of the medical record, medical record features (feat), part information (feat-part) of the medical record features, the sex of a medical record patient and the crowd attributes of the medical record patient;
3) the server encodes natural text (text) of the medical record by using CNN, encodes medical record features (feat) and part information (feat-part) of the medical record features by using an emb mode, and encodes the sex (sex) of the patient and the population attribute information (popup) of the patient by using a one-hot mode;
4) the server splices (+) various medical record information and encodes the result in the form of vector;
5) the server realizes label prediction based on a pre-constructed structured label system, and outputs a first-level part label, a second-level part label, a whole body part label and a disease category label obtained through prediction.
And in the whole process, learning of network parameters is realized by using a Sigmoid loss function.
With further reference to fig. 5, as an implementation of the method shown in the above figures, the present application provides an embodiment of an apparatus for building a label prediction model, where the embodiment of the apparatus corresponds to the embodiment of the method shown in fig. 2, and the apparatus may be applied to various electronic devices.
As shown in fig. 5, the apparatus 500 for building a label prediction model of the present embodiment may include: a disease and category label system establishing unit 501, a disease and part label system establishing unit 502, a medical record and disease label system establishing unit 503 and a label prediction model establishing unit 504. The disease and category label system establishing unit 501 is configured to establish a disease and category label system according to a first sample recorded with a correspondence between specific diseases and disease categories; a disease and part label system establishing unit 502 configured to establish a disease and part label system based on the second sample in which the correspondence between the specific disease and the part of the human body is recorded; a medical record and disease label system establishing unit 503 configured to establish a medical record and disease label system according to a third sample recorded with a correspondence between a historical medical record and a historical diagnosis disease; a label prediction model construction unit 504 configured to construct a label prediction model based on the disease and category label system, the disease and location label system, and the medical record and disease label system.
In the present embodiment, in the apparatus 500 for constructing a label prediction model: the detailed processing and the technical effects of the disease and category label system establishing unit 501, the disease and portion label system establishing unit 502, the medical record and disease label system establishing unit 503, and the label prediction model establishing unit 504 can refer to the related descriptions of step 201 and step 204 in the corresponding embodiment of fig. 2, which are not repeated herein.
In some optional implementations of this embodiment, the medical record and disease label system establishing unit 503 may be further configured to:
extracting natural language texts from historical medical record information, and determining suspected diseases according to historical diagnosis diseases;
performing feature dimension reduction processing on the natural language text to obtain dimension reduction text features;
processing the dimensionality reduction text features by using a preset neural network to obtain processed medical record features;
and establishing a label system for obtaining the medical record and the disease according to the labeled representation of the suspected disease corresponding to the processed medical record characteristics.
In some optional implementations of this embodiment, the apparatus 500 for building a label prediction model may further include:
an actual electronic medical record receiving unit configured to receive an incoming actual electronic medical record;
an actual label determination unit configured to determine an actual disease category label and an actual human body part label corresponding to an actual electronic medical record using a label prediction model;
and the actual label returning unit is configured to return the actual disease category label and the actual human body part label according to a preset path.
In some optional implementations of this embodiment, the apparatus 500 for building a label prediction model may further include:
the user attribute label determining unit is configured to determine a corresponding user attribute label according to the natural language text of the actual electronic medical record; the user attribute labels comprise a gender label, a family genetic disease label and a working environment label;
and a user attribute tag returning unit configured to return the user attribute tag as an attached reference tag.
In some optional implementations of this embodiment, the apparatus 500 for building a label prediction model may further include:
the quantity abnormal tag counting unit is configured to count quantity abnormal tags of which the actual output quantity exceeds a preset quantity in a preset time period;
and a short-term epidemic outbreak determination unit configured to determine whether there is an outbreak phenomenon of the short-term epidemic according to the quantity abnormality label.
In some optional implementations of this embodiment, the apparatus 500 for building a label prediction model may further include:
a tag error information receiving unit configured to receive tag error information incoming for a tag that is predicted to be erroneous;
a model parameter adjusting unit configured to adjust parameters of the tag prediction model according to the tag error information.
The device for constructing the label prediction model provided by the embodiment provides a construction scheme of the multi-task learning prediction model, that is, the first sample, the second sample and the third sample are simultaneously used as training samples, so that the latent commonalities and the association relations among different samples can be simultaneously learned under different task targets, the constructed multi-task learning label prediction model outputs more comprehensive and accurate prediction labels, and the technical defects that data are not shared and the accuracy of prediction results is not high in the prior art that a plurality of single-task learning models are respectively constructed and summarized are overcome.
According to an embodiment of the present application, an electronic device and a computer-readable storage medium are also provided.
FIG. 6 illustrates a block diagram of an electronic device suitable for use in implementing the method for building a label prediction model of embodiments of the present application. Electronic devices are intended to represent various forms of digital computers, such as laptops, desktops, workstations, personal digital assistants, servers, blade servers, mainframes, and other appropriate computers. The electronic device may also represent various forms of mobile devices, such as personal digital processing, cellular phones, smart phones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions, are meant to be examples only, and are not meant to limit implementations of the present application that are described and/or claimed herein.
As shown in fig. 6, the electronic apparatus includes: one or more processors 601, memory 602, and interfaces for connecting the various components, including a high-speed interface and a low-speed interface. The various components are interconnected using different buses and may be mounted on a common motherboard or in other manners as desired. The processor may process instructions for execution within the electronic device, including instructions stored in or on the memory to display graphical information of a GUI on an external input/output apparatus (such as a display device coupled to the interface). In other embodiments, multiple processors and/or multiple buses may be used, along with multiple memories and multiple memories, as desired. Also, multiple electronic devices may be connected, with each device providing portions of the necessary operations (e.g., as a server array, a group of blade servers, or a multi-processor system). In fig. 6, one processor 601 is taken as an example.
The memory 602 is a non-transitory computer readable storage medium as provided herein. Wherein the memory stores instructions executable by the at least one processor to cause the at least one processor to perform the method for building a label prediction model provided herein. A non-transitory computer readable storage medium of the present application stores computer instructions for causing a computer to perform the method for building a tag prediction model provided herein.
The memory 602 is used as a non-transitory computer readable storage medium for storing non-transitory software programs, non-transitory computer executable programs, and modules, such as program instructions/modules corresponding to the method for constructing a label prediction model in the embodiment of the present application (for example, the disease and category label system establishing unit 501, the disease and location label system establishing unit 502, the medical record and disease label system establishing unit 503, and the label prediction model establishing unit 504 shown in fig. 5). The processor 601 executes various functional applications of the server and data processing by executing non-transitory software programs, instructions and modules stored in the memory 602, namely, implements the method for constructing the tag prediction model in the above method embodiments.
The memory 602 may include a storage program area and a storage data area, wherein the storage program area may store an operating system, an application program required for at least one function; the storage data area may store various types of data and the like created by the electronic device in performing the method for constructing the tag prediction model. Further, the memory 602 may include high speed random access memory, and may also include non-transitory memory, such as at least one magnetic disk storage device, flash memory device, or other non-transitory solid state storage device. In some embodiments, the memory 602 optionally includes memory remotely located from the processor 601, and such remote memory may be connected over a network to an electronic device adapted to perform the method for building the label prediction model. Examples of such networks include, but are not limited to, the internet, intranets, local area networks, mobile communication networks, and combinations thereof.
The electronic device adapted to perform the method for building a label prediction model may further comprise: an input device 603 and an output device 604. The processor 601, the memory 602, the input device 603 and the output device 604 may be connected by a bus or other means, and fig. 6 illustrates the connection by a bus as an example.
The input device 603 may receive input numeric or character information and generate key signal inputs related to user settings and function control of an electronic apparatus suitable for performing the method for constructing the tag prediction model, such as an input device such as a touch screen, a keypad, a mouse, a track pad, a touch pad, a pointing stick, one or more mouse buttons, a track ball, a joystick, or the like. The output devices 604 may include a display device, auxiliary lighting devices (e.g., LEDs), and tactile feedback devices (e.g., vibrating motors), among others. The display device may include, but is not limited to, a Liquid Crystal Display (LCD), a Light Emitting Diode (LED) display, and a plasma display. In some implementations, the display device can be a touch screen.
Various implementations of the systems and techniques described here can be realized in digital electronic circuitry, integrated circuitry, application specific ASICs (application specific integrated circuits), computer hardware, firmware, software, and/or combinations thereof. These various embodiments may include: implemented in one or more computer programs that are executable and/or interpretable on a programmable system including at least one programmable processor, which may be special or general purpose, receiving data and instructions from, and transmitting data and instructions to, a storage system, at least one input device, and at least one output device.
These computer programs (also known as programs, software applications, or code) include machine instructions for a programmable processor, and may be implemented using high-level procedural and/or object-oriented programming languages, and/or assembly/machine languages. As used herein, the terms "machine-readable medium" and "computer-readable medium" refer to any computer program product, apparatus, and/or device (e.g., magnetic discs, optical disks, memory, Programmable Logic Devices (PLDs)) used to provide machine instructions and/or data to a programmable processor, including a machine-readable medium that receives machine instructions as a machine-readable signal. The term "machine-readable signal" refers to any signal used to provide machine instructions and/or data to a programmable processor.
To provide for interaction with a user, the systems and techniques described here can be implemented on a computer having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to a user; and a keyboard and a pointing device (e.g., a mouse or a trackball) by which a user can provide input to the computer. Other kinds of devices may also be used to provide for interaction with a user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user may be received in any form, including acoustic, speech, or tactile input.
The systems and techniques described here can be implemented in a computing system that includes a back-end component (e.g., as a data server), or that includes a middleware component (e.g., an application server), or that includes a front-end component (e.g., a user computer having a graphical user interface or a web browser through which a user can interact with an implementation of the systems and techniques described here), or any combination of such back-end, middleware, or front-end components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: local Area Networks (LANs), Wide Area Networks (WANs), and the Internet.
The computer system may include clients and servers. A client and server are generally remote from each other and typically interact through a communication network. The relationship of client and server arises by virtue of computer programs running on the respective computers and having a client-server relationship to each other.
The technical scheme of the embodiment of the application provides a construction scheme of a multi-task learning prediction model, namely, a first sample, a second sample and a third sample are used as training samples at the same time, so that the latent commonalities and incidence relations among different samples can be learned at the same time under different task targets, the constructed multi-task learning label prediction model can output more comprehensive and accurate prediction labels, and the technical defects that data are not shared and the accuracy of prediction results is not high in the prior art that a plurality of single-task learning models are obtained through respective construction and then are summarized are overcome.
It should be understood that various forms of the flows shown above may be used, with steps reordered, added, or deleted. For example, the steps described in the present application may be executed in parallel, sequentially, or in different orders, and the present invention is not limited thereto as long as the desired results of the technical solutions disclosed in the present application can be achieved.
The above-described embodiments should not be construed as limiting the scope of the present application. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions may be made in accordance with design requirements and other factors. Any modification, equivalent replacement, and improvement made within the spirit and principle of the present application shall be included in the protection scope of the present application.
Claims (14)
1. A method for constructing a label prediction model, comprising:
establishing a disease and category label system according to a first sample recorded with the corresponding relation between specific diseases and disease categories;
establishing a disease and part label system according to a second sample recorded with the corresponding relation between the specific disease and the human body part;
establishing a medical record and disease label system according to a third sample recorded with the corresponding relation between the historical medical record and the historical diagnosis disease;
and constructing a label prediction model according to the disease and category label system, the disease and part label system and the medical record and disease label system.
2. The method of claim 1, wherein the establishing a medical record and disease label system according to the third sample recorded with the corresponding relationship between the historical medical records and the historical diagnosis diseases comprises:
extracting natural language texts from the historical medical record information, and determining suspected diseases according to the historical diagnosis diseases;
performing feature dimension reduction processing on the natural language text to obtain dimension reduction text features;
processing the dimensionality reduction text features by using a preset neural network to obtain processed medical record features;
and establishing a label system for obtaining the medical record and the disease according to the labeled representation of the suspected disease corresponding to the processed medical record characteristics.
3. The method of claim 1, further comprising:
receiving an incoming actual electronic medical record;
determining an actual disease category label and an actual human body part label corresponding to the actual electronic medical record by using the label prediction model;
and returning the actual disease category label and the actual human body part label according to a preset path.
4. The method of claim 3, further comprising:
determining a corresponding user attribute label according to the natural language text of the actual electronic medical record; wherein the user attribute label comprises a gender label, a family genetic disease label and a working environment label;
returning the user attribute tag as an attached reference tag.
5. The method of claim 3 or 4, further comprising:
counting quantity abnormal labels of which the actual output quantity exceeds the preset quantity in a preset time period;
and determining whether the outbreak phenomenon of the short-term epidemic exists according to the quantity abnormal label.
6. The method of claim 1, further comprising:
receiving incoming tag error information for a tag that is mispredicted;
and adjusting parameters of the label prediction model according to the label error information.
7. An apparatus for building a label prediction model, comprising:
a disease and category label system establishing unit configured to establish a disease and category label system according to a first sample in which a correspondence relationship between a specific disease and a disease category is recorded;
a disease and part label system establishing unit configured to establish a disease and part label system based on the second sample in which the correspondence between the specific disease and the part of the human body is recorded;
a medical record and disease label system establishing unit configured to establish a medical record and disease label system according to a third sample recorded with a corresponding relationship between a historical medical record and a historical diagnosis disease;
a label prediction model construction unit configured to construct a label prediction model according to the disease and category label system, the disease and location label system, and the medical record and disease label system.
8. The apparatus of claim 7, wherein the medical record and disease label system building unit is further configured to:
extracting natural language texts from the historical medical record information, and determining suspected diseases according to the historical diagnosis diseases;
performing feature dimension reduction processing on the natural language text to obtain dimension reduction text features;
processing the dimensionality reduction text features by using a preset neural network to obtain processed medical record features;
and establishing a label system for obtaining the medical record and the disease according to the labeled representation of the suspected disease corresponding to the processed medical record characteristics.
9. The apparatus of claim 7, further comprising:
an actual electronic medical record receiving unit configured to receive an incoming actual electronic medical record;
an actual label determination unit configured to determine an actual disease category label and an actual human body part label corresponding to the actual electronic medical record using the label prediction model;
an actual tag returning unit configured to return the actual disease category tag and the actual human body part tag according to a preset path.
10. The apparatus of claim 9, further comprising:
the user attribute label determining unit is configured to determine a corresponding user attribute label according to the natural language text of the actual electronic medical record; wherein the user attribute label comprises a gender label, a family genetic disease label and a working environment label;
a user attribute tag returning unit configured to return the user attribute tag as an attached reference tag.
11. The apparatus of claim 8 or 9, further comprising:
the quantity abnormal tag counting unit is configured to count quantity abnormal tags of which the actual output quantity exceeds a preset quantity in a preset time period;
a short-term epidemic outbreak determination unit configured to determine whether there is an outbreak of a short-term epidemic according to the quantity abnormality label.
12. The apparatus of claim 7, further comprising:
a tag error information receiving unit configured to receive tag error information incoming for a tag that is predicted to be erroneous;
a model parameter adjustment unit configured to adjust a parameter of the tag prediction model according to the tag error information.
13. An electronic device, comprising:
at least one processor; and
a memory communicatively coupled to the at least one processor; wherein,
the memory stores instructions executable by the at least one processor to enable the at least one processor to perform the method for building a label prediction model of any of claims 1-6.
14. A non-transitory computer readable storage medium having stored thereon computer instructions for causing the computer to perform the method for building a tag prediction model of any one of claims 1-6.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202011022926.2A CN112117009A (en) | 2020-09-25 | 2020-09-25 | Method, device, electronic equipment and medium for constructing label prediction model |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202011022926.2A CN112117009A (en) | 2020-09-25 | 2020-09-25 | Method, device, electronic equipment and medium for constructing label prediction model |
Publications (1)
Publication Number | Publication Date |
---|---|
CN112117009A true CN112117009A (en) | 2020-12-22 |
Family
ID=73798120
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202011022926.2A Pending CN112117009A (en) | 2020-09-25 | 2020-09-25 | Method, device, electronic equipment and medium for constructing label prediction model |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN112117009A (en) |
Citations (15)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20110207618A1 (en) * | 2010-02-19 | 2011-08-25 | Industry-Academic Cooperation Foundation, Yonsei University | Class label predicting apparatus and method |
CN107845424A (en) * | 2017-11-15 | 2018-03-27 | 海南大学 | The method and system of diagnostic message Treatment Analysis |
CN108520780A (en) * | 2018-03-07 | 2018-09-11 | 中国科学院计算技术研究所 | A kind of medical data processing and system based on transfer learning |
CN109036577A (en) * | 2018-07-27 | 2018-12-18 | 合肥工业大学 | Diabetic complication analysis method and device |
CN109036553A (en) * | 2018-08-01 | 2018-12-18 | 北京理工大学 | A kind of disease forecasting method based on automatic extraction Medical Technologist's knowledge |
CN109243618A (en) * | 2018-09-12 | 2019-01-18 | 腾讯科技(深圳)有限公司 | Construction method, disease label construction method and the smart machine of medical model |
CN109545317A (en) * | 2018-10-30 | 2019-03-29 | 平安科技(深圳)有限公司 | The method and Related product of behavior in hospital are determined based on prediction model in hospital |
CN110378410A (en) * | 2019-07-16 | 2019-10-25 | 北京字节跳动网络技术有限公司 | Multi-tag scene classification method, device and electronic equipment |
CN110851738A (en) * | 2019-10-28 | 2020-02-28 | 百度在线网络技术(北京)有限公司 | Method, device and equipment for acquiring POI state information and computer storage medium |
CN110867231A (en) * | 2019-11-18 | 2020-03-06 | 中山大学 | Disease prediction method, device, computer equipment and medium based on text classification |
CN111160441A (en) * | 2019-12-24 | 2020-05-15 | 上海联影智能医疗科技有限公司 | Classification method, computer device, and storage medium |
CN111180068A (en) * | 2019-12-19 | 2020-05-19 | 浙江大学 | Chronic disease prediction system based on multi-task learning model |
KR20200062885A (en) * | 2018-11-27 | 2020-06-04 | 서강대학교산학협력단 | Device and method for dependency parsing |
CN111291564A (en) * | 2020-03-03 | 2020-06-16 | 腾讯科技(深圳)有限公司 | Model training method and device for word vector acquisition and storage medium |
CN111695053A (en) * | 2020-06-12 | 2020-09-22 | 上海智臻智能网络科技股份有限公司 | Sequence labeling method, data processing device and readable storage medium |
-
2020
- 2020-09-25 CN CN202011022926.2A patent/CN112117009A/en active Pending
Patent Citations (15)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20110207618A1 (en) * | 2010-02-19 | 2011-08-25 | Industry-Academic Cooperation Foundation, Yonsei University | Class label predicting apparatus and method |
CN107845424A (en) * | 2017-11-15 | 2018-03-27 | 海南大学 | The method and system of diagnostic message Treatment Analysis |
CN108520780A (en) * | 2018-03-07 | 2018-09-11 | 中国科学院计算技术研究所 | A kind of medical data processing and system based on transfer learning |
CN109036577A (en) * | 2018-07-27 | 2018-12-18 | 合肥工业大学 | Diabetic complication analysis method and device |
CN109036553A (en) * | 2018-08-01 | 2018-12-18 | 北京理工大学 | A kind of disease forecasting method based on automatic extraction Medical Technologist's knowledge |
CN109243618A (en) * | 2018-09-12 | 2019-01-18 | 腾讯科技(深圳)有限公司 | Construction method, disease label construction method and the smart machine of medical model |
CN109545317A (en) * | 2018-10-30 | 2019-03-29 | 平安科技(深圳)有限公司 | The method and Related product of behavior in hospital are determined based on prediction model in hospital |
KR20200062885A (en) * | 2018-11-27 | 2020-06-04 | 서강대학교산학협력단 | Device and method for dependency parsing |
CN110378410A (en) * | 2019-07-16 | 2019-10-25 | 北京字节跳动网络技术有限公司 | Multi-tag scene classification method, device and electronic equipment |
CN110851738A (en) * | 2019-10-28 | 2020-02-28 | 百度在线网络技术(北京)有限公司 | Method, device and equipment for acquiring POI state information and computer storage medium |
CN110867231A (en) * | 2019-11-18 | 2020-03-06 | 中山大学 | Disease prediction method, device, computer equipment and medium based on text classification |
CN111180068A (en) * | 2019-12-19 | 2020-05-19 | 浙江大学 | Chronic disease prediction system based on multi-task learning model |
CN111160441A (en) * | 2019-12-24 | 2020-05-15 | 上海联影智能医疗科技有限公司 | Classification method, computer device, and storage medium |
CN111291564A (en) * | 2020-03-03 | 2020-06-16 | 腾讯科技(深圳)有限公司 | Model training method and device for word vector acquisition and storage medium |
CN111695053A (en) * | 2020-06-12 | 2020-09-22 | 上海智臻智能网络科技股份有限公司 | Sequence labeling method, data processing device and readable storage medium |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN111681726B (en) | Processing method, device, equipment and medium of electronic medical record data | |
US11150875B2 (en) | Automated content editor | |
US11455981B2 (en) | Method, apparatus, and system for conflict detection and resolution for competing intent classifiers in modular conversation system | |
US11842145B1 (en) | Systems, devices, and methods for software coding | |
US10599686B1 (en) | Method and system for extracting information from graphs | |
US20180075368A1 (en) | System and Method of Advising Human Verification of Often-Confused Class Predictions | |
US12032906B2 (en) | Method, apparatus and device for quality control and storage medium | |
CN111753543A (en) | Medicine recommendation method and device, electronic equipment and storage medium | |
US20180068221A1 (en) | System and Method of Advising Human Verification of Machine-Annotated Ground Truth - High Entropy Focus | |
US20180068222A1 (en) | System and Method of Advising Human Verification of Machine-Annotated Ground Truth - Low Entropy Focus | |
CN110612522B (en) | Establishment of solid model | |
US20220147835A1 (en) | Knowledge graph construction system and knowledge graph construction method | |
US10585902B2 (en) | Cognitive computer assisted attribute acquisition through iterative disclosure | |
CN112507090B (en) | Method, apparatus, device and storage medium for outputting information | |
KR20200009117A (en) | Systems for data collection and analysis | |
JP2021099798A (en) | Method and device for structuring data, computer device, and medium | |
US20210209482A1 (en) | Method and apparatus for verifying accuracy of judgment result, electronic device and medium | |
CN110532487B (en) | Label generation method and device | |
US11080335B2 (en) | Concept-based autosuggest based on previously identified items | |
CN114078597A (en) | Decision trees with support from text for healthcare applications | |
WO2023155678A1 (en) | Method and apparatus for determining information | |
CN112015866B (en) | Method, device, electronic equipment and storage medium for generating synonymous text | |
CN112117009A (en) | Method, device, electronic equipment and medium for constructing label prediction model | |
CN112509692A (en) | Method, apparatus, electronic device and storage medium for matching medical expressions | |
US12056443B1 (en) | Apparatus and method for generating annotations for electronic records |
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 |