CN112530533B - Medical record document detection method and device and electronic equipment - Google Patents
Medical record document detection method and device and electronic equipment Download PDFInfo
- Publication number
- CN112530533B CN112530533B CN202011380398.8A CN202011380398A CN112530533B CN 112530533 B CN112530533 B CN 112530533B CN 202011380398 A CN202011380398 A CN 202011380398A CN 112530533 B CN112530533 B CN 112530533B
- Authority
- CN
- China
- Prior art keywords
- entity
- attribute information
- identity attribute
- conflict
- medical record
- 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.)
- Active
Links
- 238000001514 detection method Methods 0.000 title claims abstract description 63
- 238000000034 method Methods 0.000 claims description 35
- 230000007721 medicinal effect Effects 0.000 claims description 21
- 230000015654 memory Effects 0.000 claims description 19
- 230000005856 abnormality Effects 0.000 claims description 10
- 230000000694 effects Effects 0.000 abstract description 10
- 238000011282 treatment Methods 0.000 abstract description 9
- 238000013473 artificial intelligence Methods 0.000 abstract description 6
- 238000005516 engineering process Methods 0.000 abstract description 6
- 238000013135 deep learning Methods 0.000 abstract description 4
- 201000010099 disease Diseases 0.000 description 16
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 description 16
- 230000008569 process Effects 0.000 description 11
- 238000012549 training Methods 0.000 description 11
- 238000003745 diagnosis Methods 0.000 description 8
- 238000004590 computer program Methods 0.000 description 7
- 238000003908 quality control method Methods 0.000 description 5
- 206010020772 Hypertension Diseases 0.000 description 4
- 238000004891 communication Methods 0.000 description 4
- 208000024891 symptom Diseases 0.000 description 4
- 238000013528 artificial neural network Methods 0.000 description 3
- 238000003556 assay Methods 0.000 description 3
- 238000001914 filtration Methods 0.000 description 3
- 230000006870 function Effects 0.000 description 3
- 238000012986 modification Methods 0.000 description 3
- 230000004048 modification Effects 0.000 description 3
- 238000012552 review Methods 0.000 description 3
- 230000002159 abnormal effect Effects 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 230000018109 developmental process Effects 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 230000003993 interaction Effects 0.000 description 2
- 239000004973 liquid crystal related substance Substances 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008468 bone growth Effects 0.000 description 1
- 230000001413 cellular effect Effects 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000013527 convolutional neural network Methods 0.000 description 1
- 125000004122 cyclic group Chemical group 0.000 description 1
- 238000013136 deep learning model Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000002372 labelling Methods 0.000 description 1
- 238000010295 mobile communication Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000011218 segmentation Effects 0.000 description 1
- 230000001953 sensory effect Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
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
- G16H10/00—ICT specially adapted for the handling or processing of patient-related medical or healthcare data
- G16H10/60—ICT specially adapted for the handling or processing of patient-related medical or healthcare data for patient-specific data, e.g. for electronic patient records
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/20—Natural language analysis
- G06F40/237—Lexical tools
- G06F40/247—Thesauruses; Synonyms
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/20—Natural language analysis
- G06F40/279—Recognition of textual entities
- G06F40/289—Phrasal analysis, e.g. finite state techniques or chunking
- G06F40/295—Named entity recognition
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H50/00—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics
- G16H50/20—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for computer-aided diagnosis, e.g. based on medical expert systems
Landscapes
- Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Theoretical Computer Science (AREA)
- General Health & Medical Sciences (AREA)
- Public Health (AREA)
- Medical Informatics (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Biomedical Technology (AREA)
- Artificial Intelligence (AREA)
- Primary Health Care (AREA)
- Computational Linguistics (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Epidemiology (AREA)
- Data Mining & Analysis (AREA)
- Databases & Information Systems (AREA)
- Pathology (AREA)
- Medical Treatment And Welfare Office Work (AREA)
Abstract
The application discloses a medical record document detection method, a medical record document detection device and electronic equipment, and relates to the field of artificial intelligence such as artificial intelligence medical treatment and deep learning technology. The specific implementation scheme is as follows: acquiring at least one entity in a medical record document; identifying first identity attribute information corresponding to each entity; acquiring conflict information of first identity attribute information corresponding to the at least one entity; and carrying out anomaly detection on the medical record document according to the conflict information. According to the technology of the application, the problem that the detection effect of the medical record document detection mode is relatively poor is solved, and the detection effect of the medical record document is improved.
Description
Technical Field
The application relates to the field of artificial intelligence, in particular to the technical field of artificial intelligence medical treatment and deep learning, and specifically relates to a medical record document detection method, a medical record document detection device and electronic equipment.
Background
With the development of electronic technology, at present, an electronic medical record system is generally adopted when doctors write medical record files. Because the medical record editor in the electronic medical record system cannot simultaneously give consideration to the convenience and the collection of high-quality medical record files, various errors in the medical record files are caused, and the quality of the medical record files is caused.
At present, the quality control mode of the medical record file is usually to manually review, and quality control personnel sample the medical record file to perform quality detection or only perform form detection on the medical record file to judge the integrity of the medical record file.
Disclosure of Invention
The disclosure provides a medical record document detection method and device and electronic equipment.
According to a first aspect of the present disclosure, there is provided a medical record document detection method, including:
acquiring at least one entity in a medical record document;
identifying first identity attribute information corresponding to each entity;
acquiring conflict information of first identity attribute information corresponding to the at least one entity;
and carrying out anomaly detection on the medical record document according to the conflict information.
According to a second aspect of the present disclosure, there is provided a medical record document detection apparatus, comprising:
the first acquisition module is used for acquiring at least one entity in the medical record document;
the identification module is used for identifying first identity attribute information corresponding to each entity;
the second acquisition module is used for acquiring conflict information of the first identity attribute information corresponding to the at least one entity;
and the abnormality detection module is used for carrying out abnormality detection on the medical record document according to the conflict information.
According to a third aspect of the present disclosure, there is provided 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 any one of the methods of the first aspect.
According to a fourth aspect of the present disclosure, there is provided a non-transitory computer-readable storage medium storing computer instructions for causing a computer to perform any of the methods of the first aspect.
According to a fifth aspect of the present disclosure, there is provided a computer program product comprising a computer program which, when executed by a processor, implements any of the methods of the first aspect.
According to the technology, the problem that the detection effect of the medical record document detection mode is relatively poor is solved, and the detection effect of the medical record document is improved.
It should be understood that the description in this section is not intended to identify key or critical features of the embodiments of the disclosure, nor is it intended to be used to limit the scope of the disclosure. Other features of the present disclosure will become apparent from the following specification.
Drawings
The drawings are for better understanding of the present solution and do not constitute a limitation of the present application. Wherein:
FIG. 1 is a flow chart of a medical record document detection method according to a first embodiment of the present application;
FIG. 2 is a schematic structural view of a medical record document detection device according to a second embodiment of the present application;
FIG. 3 is a block diagram of an electronic device for implementing a medical record document detection method of an embodiment of the present application.
Detailed Description
Exemplary embodiments of the present application are described below in conjunction with the accompanying drawings, which include various details of the embodiments of the present application to facilitate understanding, and should be considered as merely exemplary. Accordingly, one of ordinary skill in the art will recognize that various changes and modifications of the embodiments described herein can be made without departing from the scope and spirit of the present application. Also, descriptions of well-known functions and constructions are omitted in the following description for clarity and conciseness.
First embodiment
As shown in fig. 1, the present application provides a medical record document detection method, which includes the following steps:
step S101: at least one entity in the medical record document is obtained.
In this embodiment, the medical record document detection method relates to an artificial intelligence technology, in particular to the technical field of artificial intelligence medical treatment and deep learning, and can be widely applied to various scenes such as clinical auxiliary decision-making systems, medical record quality control, content quality control, terminal quality control and the like. The method may be applied to an electronic device, which may be a server or a terminal, and is not specifically limited herein.
The medical record is a record of medical activity processes such as occurrence, development, regression, examination, diagnosis and treatment of a patient by medical staff, the recording modes of the medical record can be generally two, one is paper medical record, the other is electronic medical record, the electronic medical record refers to a process of recording medical activity by adopting an electronic document mode, and the electronic record refers to digitization of medical activity information.
Currently, there are various ways to electronically process medical records, for example, using text files, word documents, or scanning to electronically process medical records in an image file manner, and for example, using a unified template document editing by a medical record editor to generate a medical record document.
In this embodiment, the medical record document is an electronic document, and includes description information of a series of medical activity events such as complaints, current medical history, symptom descriptions, assays, examinations, diagnoses, and treatments of the patient, and the description information of the medical activity events may be referred to as medical record connotation.
The entity in the medical record file refers to medical activity events of the patient in the process of medical activity, including medical activity events such as complaints, current medical history, symptom descriptions, assays, examinations, diagnoses, treatments and the like, and corresponding description information of the medical activity events is recorded under each medical activity event.
That is, the medical history content may specifically include description information of a patient's complaint about a disease, description information of a patient's current medical history, description information of symptoms, description information of a disease treatment process, and the like, which are not described in detail herein.
One or more entities may be included in the medical record document, and in the case where the medical record document is relatively simple, such as where the patient is a neonate or where the patient is an onset of a disease, only one medical activity event of the patient may be recorded in the medical record document, i.e., only one entity is included in the medical record document.
In a specific implementation process, at least one entity in the medical record document can be obtained by obtaining entity words in the medical record document, wherein the entity words refer to identification of medical activity events by medical staff when recording the medical activity events, the entity words "diagnosis" refer to identification of disease diagnosis events, and the entity words "treatment" refer to identification of disease treatment events.
Because medical records typically employ semi-structured data stores, natural language text and some structured information, such as gender and age, are contained within the semi-structured data store. Thus, some entity words in the medical record document, such as complaints, medical history, diseases, symptoms, assays, examinations, and procedures, can be extracted by means and methods of natural language understanding (Natural Language Understanding, NLU) techniques, word segmentation, or entity recognition, to obtain at least one entity in the medical record document.
Accordingly, in order to increase the readability of the medical record document, the description information of the entity word and the entity usually appear synchronously, so that the description information of the corresponding entity can also be obtained at the corresponding position of the entity word, such as the position immediately following the entity word.
Step S102: first identity attribute information corresponding to each entity is identified.
The medical record file not only comprises medical record connotation, but also comprises basic information of a patient, such as identity attribute information of the patient, wherein the identity attribute information can comprise gender attribute information, age attribute information and the like.
The sex attribute information may include whether the patient is male or female, and the age attribute information may be divided into sections by the age of the patient, and may be divided into whether the patient is neonate, child, adult, or elderly. Such as a neonate when the patient is aged 0 to 1 year old, a child when the patient is aged 1 to 18 years old, an adult when the patient is aged 18 to 50 years old, and an elderly when the patient is aged 50 years old.
The purpose of the embodiment of the application is to detect the quality of the medical record connotation so as to judge whether the description information in the medical record connotation is logically reasonable, wherein the logically reasonable refers to that the description information in the medical record connotation is reasonable for the identity attribute information of a patient.
Since a medical record document is usually a record of the medical activity process of a patient, the recorded information can be related to the identity attribute information of the patient, and accordingly, the description information in the medical record connotation can have the corresponding identity attribute information. The description information in the medical record connotation is the description information of the medical activity event, so that the identity attribute information corresponding to the description information is the identity attribute information of the entity corresponding to the description information.
For example, the descriptive information of disease diagnosis is hypertension, and since hypertension is usually an adult or senile disease, the identity attribute information corresponding to the entity is adult or senile.
For another example, the descriptive information of disease diagnosis is gender-related, and thus, the entity may correspond to gender-related identity attribute information such as male or female.
In a specific implementation process, the first identity attribute information corresponding to each entity can be identified by identifying the identity attribute information of the description information corresponding to each entity. The first identity attribute information may be age attribute information, sex attribute information, or include age attribute information and sex attribute information.
The identification mode can be various, for example, the identity attribute information of the description information corresponding to each entity can be identified through a target model, and the target model can be a deep learning model constructed based on a neural network such as a convolutional neural network or a cyclic neural network.
In particular, prior to recognition, a training data set may be constructed to train the target model, which may include training input samples and training output samples.
The training input sample may include an entity list of a plurality of medical records, denoted as e= [ E ] 1 ，e 2 ，…，e m ]，e i And (3) representing an entity list of the ith medical record document, wherein m is a positive integer greater than 1.
Entity list e i Can be marked as e i ＝[w 1 ，w 2 ，…，w n ]All entity information in the ith medical record file is included in the medical record file, and the entity can be all the entity after the duplication removal. Wherein w is j The j-th entity information in the i-th medical record document can be included, the entity information can include entity words and description information corresponding to the entity words, and n is a positive integer greater than 1.
The training output sample can be a label of identity attribute information corresponding to the entity word, the label can be male or female when the identity attribute information is gender attribute information, and the label can be neonate, child, adult or elderly when the identity attribute information is age attribute information.
The training input sample can be input into the target model to obtain the output of the model, the loss of the label corresponding to the entity word and the output of the model are calculated, and the parameters of the neural network are updated through a gradient descent algorithm. And obtaining trained model parameters through multiple rounds of iteration, and completing the training of the target model at the moment.
After training is completed, the first identity attribute information corresponding to each entity in the medical record document can be predicted through the target model. Descriptive information of each entity word can be input into the target model to obtain first identity attribute information corresponding to the entity word.
If training is based on gender attribute information, gender attribute information corresponding to each entity is obtained based on the target model, and if training is based on age attribute information, age attribute information corresponding to each entity is obtained based on the target model. Of course, in the case that the first identity attribute information includes gender attribute information and age attribute information, two target models may be used for training respectively, and the gender attribute information and the age attribute information corresponding to each entity may be obtained based on the trained two target models.
For another example, the first identity attribute information corresponding to each entity may be determined by matching the description information corresponding to the entity word with a pre-constructed database.
Specifically, the pre-built database may store the target word and the identity attribute information in an associated manner, for example, the target word "hypertension" may store the identity attribute information "elderly" in an associated manner, and for example, the target word "andrology" may store the identity attribute information "men" in an associated manner.
And if the description information corresponding to the entity word comprises the target word in the database or comprises a word similar to the target word in the database, determining that the description information corresponding to the entity word is successfully matched, correspondingly acquiring identity attribute information associated with the successfully matched target word, and taking the identity attribute information as first identity attribute information corresponding to the entity.
Step S103: and acquiring conflict information of the first identity attribute information corresponding to the at least one entity.
In general, in the case that the medical record document is of normal quality, the first identity attribute information corresponding to the at least one entity in the medical record connotation is generally consistent with the second identity attribute information of the patient. On the premise of the application, whether the related description in the medical record connotation is logically reasonable or not can be judged by acquiring conflict information of the first identity attribute information corresponding to the at least one entity.
The conflict information may include only first conflict information between the first identity attribute information corresponding to the at least one entity and second identity attribute information of the patient, may include only second conflict information between the first identity attribute information corresponding to the at least one entity, or may include both the first conflict information and the second conflict information.
The first conflict information is used for representing conflict situations between first identity attribute information corresponding to the at least one entity and second identity attribute information of the patient. And in the case that the first identity attribute information corresponding to one or more entities in the at least one entity is different from the second identity attribute information of the patient, indicating that conflict exists between the first identity attribute information corresponding to the at least one entity and the second identity attribute information of the patient. And under the condition that the first identity attribute information corresponding to all the entities in the at least one entity is the same as the second identity attribute information of the patient, indicating that no conflict exists between the first identity attribute information corresponding to the at least one entity and the second identity attribute information of the patient.
For example, the identity attribute information corresponding to the entity word "disease diagnosis" is male, and the identity attribute information of the patient of the medical record document is female, so that it is known that the identity attribute information corresponding to the entity word is different from the identity attribute information of the patient, a conflict exists between the first identity attribute information corresponding to the entity word and the second identity attribute information of the patient, and the description information corresponding to the entity word has a logic problem.
The second conflict information is used for representing conflict situations among first identity attribute information corresponding to the at least one entity. And under the condition that the first identity attribute information corresponding to two or more entities is different in the at least one entity, indicating that conflict exists between the first identity attribute information corresponding to the at least one entity. And under the condition that the first identity attribute information corresponding to all the entities in the at least one entity is the same, indicating that no conflict exists between the first identity attribute information corresponding to the at least one entity.
For example, the identity attribute information corresponding to the entity word "disease diagnosis" is old, that is, the description information of disease diagnosis indicates that the disease of the patient is a disease of the old, such as hypertension, and the identity attribute information corresponding to the entity word "disease examination" is child, that is, the description information of disease examination indicates that the examination made by the patient is an examination of child, such as an examination of bone growth of child. It can be known that the first identity attribute information corresponding to the two entities is different, which indicates that there is no conflict between the first identity attribute information corresponding to the at least one entity.
The conflict information may store entity words having conflicts in the at least one entity in a list manner, and may be stored in one or more lists. In the case where the conflict information includes only the first conflict information or the second conflict information, one list may be used to store all entity words in which there is a conflict in the at least one entity, and in the case where the conflict information includes both the first conflict information and the second conflict information, two lists may be used to store all entity words in the first conflict information and all entity words in the second conflict information, respectively. Of course, in case there is no conflicting entity in the at least one entity, the list may be empty.
In addition, the conflict information can be obtained by judging whether a conflict exists between the first identity attribute information corresponding to the at least one entity and obtaining entity words with the conflict, and/or judging whether a conflict exists between the first identity attribute information corresponding to the at least one entity and the second identity attribute information and obtaining the entity words with the conflict.
Of course, in the case where there is no conflicting entity word, the list to which the conflicting information corresponds may be empty, or the conflicting information may be characterized by zero.
Step S104: and carrying out anomaly detection on the medical record document according to the conflict information.
And determining that the related description in the medical record connotation has a logic problem under the condition that the conflict information represents that the conflict exists between the first identity attribute information corresponding to the at least one entity and/or the conflict information represents that the first identity attribute information corresponding to the at least one entity conflicts with the second identity attribute information of the patient.
In an optional implementation manner, whether an entity word exists in the list corresponding to the conflict information can be detected, and if the entity word exists, it is determined that a logic problem exists in the related description in the traversal meaning, the medical record document is abnormal, and it can be determined that the logic problem exists in the description information corresponding to the entity word in the medical record meaning. And under the condition that no entity word exists, determining that the medical record document is normal.
In this embodiment, at least one entity in the medical record document is obtained; identifying first identity attribute information corresponding to each entity; acquiring conflict information of first identity attribute information corresponding to the at least one entity; and carrying out anomaly detection on the medical record document according to the conflict information. Therefore, the logical rationality detection can be automatically carried out on the medical record connotations in all medical record documents in the dimension of the gender attribute and/or the age attribute, and compared with the manual review and the form detection of the content in the medical record documents, the detection effect of the medical record documents can be greatly improved.
Optionally, the step S102 specifically includes:
inputting a target entity into a target model for probability prediction so as to output M probabilities of the target entity, wherein the M probabilities are probabilities of M preset identity attribute information corresponding to the target entity respectively, and M is a positive integer greater than 1;
and under the condition that different probability values exist in the M probabilities and the maximum probability value is larger than a preset threshold, determining preset identity attribute information corresponding to the maximum probability value as first identity attribute information corresponding to the target entity, wherein the target entity is any entity in the at least one entity.
In this embodiment, for each target entity in the at least one entity, the target entity may be input to a pre-trained target model to perform probability prediction, where the target model may output M probabilities of the target entity, where the M probabilities are probabilities of M preset identity attribute information corresponding to the target entity, respectively.
The value of M may be determined according to the number of preset identity attribute information, for example, in the case where the identity attribute information is sex attribute information, since the number of preset identity attribute information is 2, that is, male and female, the target model may output two probabilities, that is, the probability that the target entity corresponds to male and the probability that the target entity corresponds to female, respectively.
For another example, in the case that the identity attribute information is age attribute information, since the number of preset identity attribute information is 4, that is, neonates, children, adults, and elderly, the target model may output four probabilities, which are the probability that the target entity corresponds to the neonate, the probability that the target entity corresponds to the child, the probability that the target entity corresponds to the adult, and the probability that the target entity corresponds to the elderly, respectively.
And under the condition that different probability values exist in the M probabilities and the maximum probability value is larger than a preset threshold value, determining preset identity attribute information corresponding to the maximum probability value as first identity attribute information corresponding to the target entity.
And determining that the target entity is identity attribute information under the condition that probability values in the M probabilities are the same or the maximum probability value in the M probabilities is smaller than or equal to a preset threshold value.
For example, object model output o 1 ，o 2 ]，o 1 Representing the probability of the target entity corresponding to a male, o 2 Representing the probability that the target entity corresponds to a female. o (o) 1 Greater than o 2 And o is o 1 When the first identity attribute information is more than 0.9, the first identity attribute information corresponding to the target entity is male, o 2 Greater than o 1 And o is o 2 When the sex attribute information is more than 0.9, the first identity attribute information corresponding to the target entity is female, and other conditions are that the target entity does not have the corresponding sex attribute.
For another example, the object model outputs [ p ] 1 ，p 2 ，p 3 ，p 4 ]，p 1 Representing the probability of the target entity corresponding to the neonate, p 2 Representing the probability of the target entity corresponding to the child, p 3 Representing the probability of the target entity corresponding to adulthood, p 4 Representing the probability that the target entity corresponds to elderly people. At p 1 When the first identity attribute information is the maximum value and is more than 0.9, the first identity attribute information corresponding to the target entity is a neonate, and p is as follows 2 Under the condition that the first identity attribute information is the maximum value and is more than 0.9, the first identity attribute information corresponding to the target entity is children, and the first identity attribute information is p 3 In the case of maximum value and greater than 0.9, the first identity attribute information corresponding to the target entity is adult, at p 4 And under the condition that the first identity attribute information is the maximum value and is more than 0.9, the first identity attribute information corresponding to the target entity is old, and other target entities have no corresponding age attribute.
In this embodiment, based on the deep learning technology, the first identity attribute information corresponding to the entity in the medical record document is identified, so that labeling of the identity attribute information corresponding to the entity in the medical record document can be avoided, thereby saving human resources and cost, and improving the detection coverage rate of the medical record document.
Optionally, before the step S103, the method further includes:
acquiring second identity attribute information of a patient of the medical record document;
the step S103 specifically includes:
and generating first conflict information comprising conflict entities under the condition that the conflict entities exist in the at least one entity, wherein the conflict entities are entities with corresponding first identity attribute information different from the second identity attribute information.
In this embodiment, the conflict information may include first conflict information, where the first conflict information may be conflict information between first identity attribute information corresponding to the at least one entity and second identity attribute information of the patient, so that before step S103, the second identity attribute information of the patient of the medical record document needs to be acquired.
Specifically, the basic information of the patient can be analyzed from the structured data in the medical record document, and the relevant information of the sex and the age of the patient can be extracted. The sex attribute information of the patient may be directly obtained from the sex of the patient, and the age attribute information of the patient may be divided into sections by the age of the patient, and may be divided into neonates, children, adults, or elderly. Such as a neonate when the patient is aged 0 to 1 year old, a child when the patient is aged 1 to 18 years old, an adult when the patient is aged 18 to 50 years old, and an elderly when the patient is aged 50 years old.
And then, judging whether the first identity attribute information of each entity is the same as the second identity attribute information, and determining that the entity is a conflict entity and adding the conflict entity into the first conflict information under different conditions until the judgment of all the entities in the medical record document is completed.
In this embodiment, the conflict information between the first identity attribute information corresponding to the at least one entity and the second identity attribute information of the patient is obtained, and the abnormality detection is performed on the medical record document according to the conflict information. Therefore, the entity which conflicts with the second identity attribute information of the patient in the medical record document can be detected, so that whether the medical record meaning in the medical record document is reasonable or not can be logically judged, and the detection effect of the medical record document can be improved.
Optionally, the step S103 specifically includes:
and generating second conflict information comprising the conflict entity combination under the condition that the medical record document has the conflict entity combination, wherein the conflict entity combination is a combination of entities with different corresponding first identity attribute information.
In this embodiment, the conflict information may include only the second conflict information, or include the first conflict entity and also include the second conflict entity, where the second conflict information may be conflict information between the first identity attribute information corresponding to the at least one entity.
In this embodiment, it may be determined whether the first identity attribute information of each entity combination is the same, and if the first identity attribute information of the entities in the entity combination is all different, the corresponding combination of the entities with different first identity attribute information is determined to be a conflicting entity combination, and the conflicting entity combination is added to the second conflicting information until the determination of all the entity combinations in the medical record document is completed. Wherein the entity combination is any two or more entities.
In this embodiment, the conflict information between the first identity attribute information corresponding to the at least one entity is obtained, and the abnormality detection is performed on the medical record document according to the conflict information. Therefore, the entity combination with conflict between the corresponding first identity attribute information in the medical record document can be detected, namely, the description information of the entities in the medical record connotation is abnormal, such as the conflict between the description information of the main complaint and the description information of the current medical history, so that whether the medical record connotation in the medical record document is reasonable can be logically judged, and the detection effect of the medical record document can be improved.
Optionally, before the step S104, the method further includes:
determining a first location of the conflicting entity in the medical record document;
and deleting the conflict entity from the first conflict information when a first object associated with the patient exists at a second position of the medical record document and the identity attribute information of the first object is the same as the first identity attribute information corresponding to the conflict entity.
In this embodiment, the description information corresponding to the conflict entity in the first conflict information may not be a description related to the patient, but may be a description related to the first object associated with the patient, such as a spouse, a mother, a child, and the like of the patient.
In the case where the identity attribute information of the patient is different from the identity attribute information of the first object, the first identity attribute information corresponding to the conflicting entity may be different from the second identity attribute information of the patient, in which case the logic of the medical record connotation is reasonable. Thus, for each conflicting entity in the first conflicting information, a double body filtering may be performed.
Specifically, the conflicting entity can be determined to be at a first location of the medical record document, and a determination can be made as to whether a first object associated with the patient exists at a second location of the medical record document. The second position of the medical record document can be a position near the first position, such as a position where the text gap cannot be separated by a preset threshold value, or a position in the same line as the conflict entity word.
Whether the first object associated with the patient is present at the second location of the medical record document may be determined by searching for a subject word, such as wife, husband, mother, son, or the like, at the second location. If a first object associated with the patient exists, and the identity attribute information of the first object is clear, if the sex attribute information of the husband is male and the sex attribute information of the mother is female, and the identity attribute information of the first object is the same as the first identity attribute information corresponding to the conflict entity, the description information of the conflict entity is logically reasonable, and the conflict entity can be deleted from the first conflict information.
The same two-body filtering manner can be adopted for the conflict entity combination in the second conflict information, so that the conflict entity combination in the second conflict information can be filtered. For example, if a main word exists near the positions of two entities in the conflict entity combination, for example, if a main word exists near the position of one entity, the main word is a husband, the identity attribute information of the object is the same as the first identity attribute information corresponding to the other entity in the conflict entity combination, and the first identity attribute information corresponding to the entity is the same as the second identity attribute information of the patient, it is indicated that the description information of the conflict entity combination is logically reasonable, and the conflict entity combination can be deleted from the second conflict information.
And then, performing anomaly detection on the medical record document according to the filtered first conflict information and/or second conflict information.
In this embodiment, the first conflict information and/or the second conflict information are filtered by the double-body filtering manner, and then the medical record document can be detected abnormally according to the filtered first conflict information and/or second conflict information, so that the conflict entity and conflict entity combination with the logic of medical record connotation being substantially reasonable can be deleted, the interference of the conflict entity and conflict entity combination in anomaly detection on the medical record document is avoided, and the detection effect of the medical record document is further improved.
Second embodiment
As shown in fig. 2, the present application provides a medical record document detection apparatus 200, including:
a first obtaining module 201, configured to obtain at least one entity in a medical record document;
an identifying module 202, configured to identify first identity attribute information corresponding to each entity;
a second obtaining module 203, configured to obtain conflict information of the first identity attribute information corresponding to the at least one entity;
and the abnormality detection module 204 is configured to perform abnormality detection on the medical record document according to the conflict information.
Optionally, the identifying module 202 includes:
The probability prediction unit is used for inputting a target entity into a target model to perform probability prediction so as to output M probabilities of the target entity, wherein the M probabilities are probabilities of M preset identity attribute information corresponding to the target entity respectively, and the M is a positive integer greater than 1;
and the determining unit is used for determining preset identity attribute information corresponding to the maximum probability value as first identity attribute information corresponding to the target entity when different probability values exist in the M probabilities and the maximum probability value is larger than a preset threshold, wherein the target entity is any entity in the at least one entity.
Optionally, the method further comprises:
a third obtaining module, configured to obtain second identity attribute information of a patient of the medical record document;
the second obtaining module 203 includes:
the first generation unit is configured to generate, when a conflict entity exists in the at least one entity, first conflict information including the conflict entity, where the conflict entity is an entity whose corresponding first identity attribute information is different from the second identity attribute information.
Optionally, the second obtaining module 203 includes:
And the second generation unit is used for generating second conflict information comprising the conflict entity combination under the condition that the medical record document has the conflict entity combination, wherein the conflict entity combination is a combination of entities with different corresponding first identity attribute information.
Optionally, the method further comprises:
a determining module, configured to determine a first location of the conflicting entity in the medical record document;
and the deleting module is used for deleting the conflict entity from the first conflict information when a first object associated with the patient exists at the second position of the medical record document and the identity attribute information of the first object is the same as the first identity attribute information corresponding to the conflict entity.
The medical record document detection device 200 provided by the application can realize each process realized by the embodiment of the medical record document detection method, and can achieve the same beneficial effects, so that repetition is avoided, and the description is omitted here.
According to embodiments of the present application, there is also provided an electronic device, a computer program product, and a readable storage medium.
As shown in fig. 3, a block diagram of an electronic device of a medical record document detection method according to an embodiment of the present application is shown. 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 telephones, smartphones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions, are meant to be exemplary only, and are not meant to limit implementations of the application described and/or claimed herein.
As shown in fig. 3, the electronic device includes: one or more processors 301, memory 302, and interfaces for connecting the various components, including high-speed interfaces and low-speed interfaces. 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 executing within the electronic device, including instructions stored in or on memory to display graphical information of the GUI on an external input/output device, such as a display device coupled to the interface. In other embodiments, multiple processors and/or multiple buses may be used, if desired, along with multiple memories and multiple memories. Also, multiple electronic devices may be connected, each providing a portion of the necessary operations (e.g., as a server array, a set of blade servers, or a multiprocessor system). One processor 301 is illustrated in fig. 3.
Memory 302 is a non-transitory computer-readable storage medium provided herein. The memory stores instructions executable by the at least one processor to cause the at least one processor to perform the medical record document detection methods provided herein. The non-transitory computer readable storage medium of the present application stores computer instructions for causing a computer to perform the medical record document detection method provided by the present application.
The memory 302 is used as a non-transitory computer readable storage medium, and can be used to store non-transitory software programs, non-transitory computer executable programs, and modules, such as program instructions/modules (e.g., the first acquisition module 201, the identification module 202, the second acquisition module 203, and the abnormality detection module 204 shown in fig. 2) corresponding to the medical record document detection method in the embodiment of the present application. The processor 301 executes the non-transitory software programs, instructions, and modules stored in the memory 302 to perform various functional applications and data processing of the server, i.e., to implement the medical record document detection method in the above-described method embodiments.
Memory 302 may include a storage program area that may store an operating system, at least one application program required for functionality, and a storage data area; the storage data area may store data created by use of an electronic device according to the method of an embodiment of the present application, and the like. In addition, memory 302 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 302 can optionally include memory remotely located relative to the processor 301, which can be coupled to the electronic device of the medical record document detection method via a network. 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 of the method of the embodiment of the application may further include: an input device 303 and an output device 304. The processor 301, memory 302, input device 303, and output device 304 may be connected by a bus or other means, for example in fig. 3.
The input device 303 may receive input numeric or character information and generate key signal inputs related to user settings and function control of the electronic device of the methods of embodiments of the present application, such as a touch screen, a keypad, a mouse, a track pad, a touch pad, a pointer stick, one or more mouse buttons, a track ball, a joystick, and the like. The output device 304 may include a display apparatus, auxiliary lighting devices (e.g., LEDs), haptic feedback devices (e.g., vibration motors), and the like. 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 may be a touch screen.
Various implementations of the systems and techniques described here can be realized in digital electronic circuitry, integrated circuitry, application specific ASIC (application specific integrated circuit), computer hardware, firmware, software, and/or combinations thereof. These various embodiments may include: implemented in one or more computer programs, the one or more computer programs may be executed and/or interpreted on a programmable system including at least one programmable processor, which may be a special purpose or general-purpose programmable processor, that may receive data and instructions from, and transmit data and instructions to, a storage system, at least one input device, and at least one output device.
These computing programs (also referred to as programs, software applications, or code) include machine instructions for a programmable processor, and may be implemented in a high-level procedural and/or object-oriented programming language, and/or in assembly/machine language. 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 pointing device (e.g., a mouse or 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 may 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 input, speech input, or tactile input.
The systems and techniques described here can be implemented in a computing system that includes a background 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 background, 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 a client and a server. The client and server are typically 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 server can be a cloud server, also called a cloud computing server or a cloud host, and is a host product in a cloud computing service system, so that the defects of high management difficulty and weak service expansibility in the traditional physical hosts and VPS service ("Virtual Private Server" or simply "VPS") are overcome.
In this embodiment, at least one entity in the medical record document is obtained; identifying first identity attribute information corresponding to each entity; acquiring conflict information of first identity attribute information corresponding to the at least one entity; and carrying out anomaly detection on the medical record document according to the conflict information. Therefore, the logical rationality detection can be automatically carried out on the medical record connotations in all medical record documents in the dimension of the gender attribute and/or the age attribute, and compared with the manual review and the form detection of the content in the medical record documents, the detection effect of the medical record documents can be greatly improved. Therefore, according to the technical scheme of the embodiment of the application, the problem that the detection effect of the medical record document detection mode is relatively poor is well solved.
It should be appreciated that various forms of the flows shown above may be used to reorder, add, or delete steps. For example, the steps described in the present application may be performed in parallel, sequentially, or in a different order, provided that the desired results of the technical solutions disclosed in the present application can be achieved, and are not limited herein.
The above embodiments do not limit the scope of the application. It will be apparent to those skilled in the art that various modifications, combinations, sub-combinations and alternatives are possible, depending on design requirements and other factors. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present application are intended to be included within the scope of the present application.
Claims (8)
1. A medical record document detection method, comprising:
acquiring at least one entity in a medical record document; an entity refers to a patient's medical activity event during a medical activity;
identifying first identity attribute information corresponding to each entity; the first identity attribute information is identity attribute information indicated by description information of a medical activity event corresponding to the entity;
acquiring conflict information of first identity attribute information corresponding to the at least one entity; the conflict information comprises first conflict information between first identity attribute information corresponding to the at least one entity and second identity attribute information of a patient of the medical record document, the first conflict information comprises conflict entities, and the conflict entities are entities with different corresponding first identity attribute information and second identity attribute information in the at least one entity; the second identity attribute information comprises gender information and age information of the patient;
performing anomaly detection on the medical record document according to the conflict information;
before the abnormality detection is performed on the medical record document according to the conflict information, the method further comprises:
determining a first location of the conflicting entity in the medical record document;
Deleting the conflict entity from the first conflict information when a first object associated with the patient exists at a second position of the medical record document and the identity attribute information of the first object is the same as the first identity attribute information corresponding to the conflict entity; the second position is a position near the first position.
2. The method of claim 1, wherein the identifying the first identity attribute information corresponding to each entity comprises:
inputting a target entity into a target model for probability prediction so as to output M probabilities of the target entity, wherein the M probabilities are probabilities of M preset identity attribute information corresponding to the target entity respectively, and M is a positive integer greater than 1;
and under the condition that different probability values exist in the M probabilities and the maximum probability value is larger than a preset threshold, determining preset identity attribute information corresponding to the maximum probability value as first identity attribute information corresponding to the target entity, wherein the target entity is any entity in the at least one entity.
3. The method according to any one of claims 1 to 2, wherein the obtaining conflict information of the first identity attribute information corresponding to the at least one entity includes:
And generating second conflict information comprising the conflict entity combination under the condition that the medical record document has the conflict entity combination, wherein the conflict entity combination is a combination of entities with different corresponding first identity attribute information.
4. A medical record document detection device, comprising:
the first acquisition module is used for acquiring at least one entity in the medical record document; an entity refers to a patient's medical activity event during a medical activity;
the identification module is used for identifying first identity attribute information corresponding to each entity; the first identity attribute information is identity attribute information indicated by description information of a medical activity event corresponding to the entity;
the second acquisition module is used for acquiring conflict information of the first identity attribute information corresponding to the at least one entity; the conflict information comprises first conflict information between first identity attribute information corresponding to the at least one entity and second identity attribute information of a patient of the medical record document, the first conflict information comprises conflict entities, and the conflict entities are entities with different corresponding first identity attribute information and second identity attribute information in the at least one entity; the second identity attribute information comprises gender information and age information of the patient;
The abnormality detection module is used for carrying out abnormality detection on the medical record document according to the conflict information;
the apparatus further comprises:
a determining module, configured to determine a first location of the conflicting entity in the medical record document;
a deleting module, configured to delete the conflicting entity from the first conflicting information when a first object associated with the patient exists at a second location of the medical record document and identity attribute information of the first object is the same as first identity attribute information corresponding to the conflicting entity; the second position is a position near the first position.
5. The apparatus of claim 4, wherein the identification module comprises:
the probability prediction unit is used for inputting a target entity into a target model to perform probability prediction so as to output M probabilities of the target entity, wherein the M probabilities are probabilities of M preset identity attribute information corresponding to the target entity respectively, and M is a positive integer greater than 1;
and the determining unit is used for determining preset identity attribute information corresponding to the maximum probability value as first identity attribute information corresponding to the target entity when different probability values exist in the M probabilities and the maximum probability value is larger than a preset threshold, wherein the target entity is any entity in the at least one entity.
6. The apparatus of any of claims 4 to 5, wherein the second acquisition module comprises:
and the second generation unit is used for generating second conflict information comprising the conflict entity combination under the condition that the medical record document has the conflict entity combination, wherein the conflict entity combination is a combination of entities with different corresponding first identity attribute information.
7. 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 of any one of claims 1-3.
8. A non-transitory computer readable storage medium storing computer instructions for causing the computer to perform the method of any one of claims 1-3.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011380398.8A CN112530533B (en) | 2020-11-30 | 2020-11-30 | Medical record document detection method and device and electronic equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011380398.8A CN112530533B (en) | 2020-11-30 | 2020-11-30 | Medical record document detection method and device and electronic equipment |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112530533A CN112530533A (en) | 2021-03-19 |
| CN112530533B true CN112530533B (en) | 2024-04-02 |
Family
ID=74995634
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011380398.8A Active CN112530533B (en) | 2020-11-30 | 2020-11-30 | Medical record document detection method and device and electronic equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112530533B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116646044B (en) * | 2023-05-16 | 2024-03-26 | 上海景栗信息科技有限公司 | Active management method and system for tumor patients based on WeChat ecological cycle |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2007148510A (en) * | 2005-11-24 | 2007-06-14 | Toshiba Corp | Medical information management system |
| CN101408875A (en) * | 2007-10-12 | 2009-04-15 | 株式会社理光 | Information processing device, image forming device, program and document data forming method |
| CN106682219A (en) * | 2017-01-03 | 2017-05-17 | 腾讯科技（深圳）有限公司 | Association document acquisition method and device |
| WO2019076114A1 (en) * | 2017-10-20 | 2019-04-25 | 阿里巴巴集团控股有限公司 | Document verification and identity verification method and device |
| CN110751994A (en) * | 2019-08-28 | 2020-02-04 | 云知声智能科技股份有限公司 | Differential display method and system for medical record template |
| CN111341454A (en) * | 2018-12-19 | 2020-06-26 | 中国电信股份有限公司 | Data mining method and device |
| CN111339314A (en) * | 2020-02-19 | 2020-06-26 | 北京百度网讯科技有限公司 | Method and device for generating triple-group data and electronic equipment |
| CN111767707A (en) * | 2020-06-30 | 2020-10-13 | 平安科技（深圳）有限公司 | Method, device, equipment and storage medium for detecting Rayleigh case |
| CN111832298A (en) * | 2020-07-14 | 2020-10-27 | 北京百度网讯科技有限公司 | Quality inspection method, device and equipment for medical records and storage medium |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20170270250A1 (en) * | 2016-03-21 | 2017-09-21 | International Business Machines Corporation | Building a patient's medical history from disparate information sources |
-
2020
- 2020-11-30 CN CN202011380398.8A patent/CN112530533B/en active Active
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2007148510A (en) * | 2005-11-24 | 2007-06-14 | Toshiba Corp | Medical information management system |
| CN101408875A (en) * | 2007-10-12 | 2009-04-15 | 株式会社理光 | Information processing device, image forming device, program and document data forming method |
| CN106682219A (en) * | 2017-01-03 | 2017-05-17 | 腾讯科技（深圳）有限公司 | Association document acquisition method and device |
| WO2019076114A1 (en) * | 2017-10-20 | 2019-04-25 | 阿里巴巴集团控股有限公司 | Document verification and identity verification method and device |
| CN111341454A (en) * | 2018-12-19 | 2020-06-26 | 中国电信股份有限公司 | Data mining method and device |
| CN110751994A (en) * | 2019-08-28 | 2020-02-04 | 云知声智能科技股份有限公司 | Differential display method and system for medical record template |
| CN111339314A (en) * | 2020-02-19 | 2020-06-26 | 北京百度网讯科技有限公司 | Method and device for generating triple-group data and electronic equipment |
| CN111767707A (en) * | 2020-06-30 | 2020-10-13 | 平安科技（深圳）有限公司 | Method, device, equipment and storage medium for detecting Rayleigh case |
| CN111832298A (en) * | 2020-07-14 | 2020-10-27 | 北京百度网讯科技有限公司 | Quality inspection method, device and equipment for medical records and storage medium |
Non-Patent Citations (2)
| Title |
|---|
| Idir Amine Amarouche 等.Electronic Health Record Data-as-a-Services Composition Based on Query Rewriting.2. Transactions on Large-Scale Data- and Knowledge-Centered Systems IV.2011,全文. * |
| 基于大数据HBase的电子病历智能全文检索系统研究;黄晓琴;戴静娟;徐海东;;中国数字医学(05);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112530533A (en) | 2021-03-19 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111292821B (en) | Medical diagnosis and treatment system | |
| CN111681726B (en) | Processing method, device, equipment and medium of electronic medical record data | |
| US11410758B2 (en) | Medical scan artifact detection system | |
| US20220076810A1 (en) | Medical scan interface feature evaluating system and methods for use therewith | |
| US11881293B2 (en) | Methods for automatic cohort selection in epidemiologic studies and clinical trials | |
| US20200051238A1 (en) | Anatomical Segmentation Identifying Modes and Viewpoints with Deep Learning Across Modalities | |
| CN112507701B (en) | Identification method, device, equipment and storage medium of medical data to be corrected | |
| CN111832298B (en) | Medical record quality inspection method, device, equipment and storage medium | |
| CN111033632A (en) | Collapsing clinical event data into meaningful patient care states | |
| CN107391933A (en) | A kind of electronic health record quality control management device and method | |
| CN109191451B (en) | Abnormality detection method, apparatus, device, and medium | |
| US20220172809A9 (en) | Report generating system and methods for use therewith | |
| US11847411B2 (en) | Obtaining supported decision trees from text for medical health applications | |
| CN112541066B (en) | Text-structured-based medical and technical report detection method and related equipment | |
| CN111383754B (en) | Medical decision method, medical decision device, electronic device, and storage medium | |
| CN110648318A (en) | Auxiliary analysis method and device for skin diseases, electronic equipment and storage medium | |
| CN112530533B (en) | Medical record document detection method and device and electronic equipment | |
| Ghosh et al. | A quantitative lung computed tomography image feature for multi-center severity assessment of COVID-19 | |
| CN112509691B (en) | Method and device for prompting differential diagnosis, electronic equipment and storage medium | |
| CN113838573B (en) | Clinical assistant decision-making diagnosis self-learning method, device, equipment and storage medium | |
| Ren et al. | A contrastive predictive coding-based classification framework for healthcare sensor data | |
| CN112507722B (en) | Method and device for controlling content quality of electronic medical record | |
| US10839299B2 (en) | Non-leading computer aided detection of features of interest in imagery | |
| Fan et al. | Positive-aware lesion detection network with cross-scale feature pyramid for OCT images | |
| US11379980B2 (en) | Image processing method, apparatus, electronic device and storage medium |
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 |