CN111696637A - Quality detection method and related device for medical record data - Google Patents
Quality detection method and related device for medical record data Download PDFInfo
- Publication number
- CN111696637A CN111696637A CN202010416797.9A CN202010416797A CN111696637A CN 111696637 A CN111696637 A CN 111696637A CN 202010416797 A CN202010416797 A CN 202010416797A CN 111696637 A CN111696637 A CN 111696637A
- Authority
- CN
- China
- Prior art keywords
- feature vector
- sample
- medical record
- record data
- anchor
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- 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
- 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)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Biomedical Technology (AREA)
- General Physics & Mathematics (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Artificial Intelligence (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Biophysics (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Life Sciences & Earth Sciences (AREA)
- Epidemiology (AREA)
- Medical Informatics (AREA)
- Primary Health Care (AREA)
- Public Health (AREA)
- Medical Treatment And Welfare Office Work (AREA)
Abstract
The application relates to intelligent decision making in artificial intelligence, and discloses a quality detection method of medical record data and a related device, wherein the method comprises the following steps: acquiring medical record data to be subjected to quality detection; vectorizing medical record data to be subjected to quality detection to obtain a first feature vector; acquiring a first anchor sample feature vector set corresponding to the anchor sample medical record data set; inputting the first anchor sample feature vector set into a trained generator to obtain a plurality of second anchor sample feature vectors; determining a first anchor sample average feature vector according to the second anchor sample feature vector; performing vector operation on the first feature vector and the first anchor sample average feature vector to obtain a second feature vector; and inputting the second feature vector into the trained discriminator to obtain a quality detection result. The scheme also relates to a block chain technology, and can be applied to the field of smart medical treatment, so that the construction of a smart city is promoted.
Description
Technical Field
The present application relates to the field of computer technologies, and in particular, to a method and a related apparatus for quality detection of medical record data.
Background
With the rapid development of information technology, the medical industry has also advanced a new stage of development. Nowadays, electronic medical record systems are already popular in most hospitals, and filling electronic medical records in computers has gradually replaced manual writing of medical records.
Generally, no matter the medical records are written manually or filled by an electronic medical record system, doctors are required to check medical record data, so that the quality of the medical record data is controlled in real time. That is, in the prior art, quality control of medical record data is completed manually, and this quality detection method is inefficient and cannot reflect quality of medical record data in time.
Disclosure of Invention
The embodiment of the application provides a quality detection method and a related device for medical record data, and by implementing the embodiment of the application, the quality detection efficiency of the medical record data is improved, and the quality condition of the medical record data can be reflected in time.
The first aspect of the present application provides a method for detecting quality of medical record data, including:
acquiring medical record data to be subjected to quality detection;
vectorizing the medical record data to be subjected to quality detection to obtain a first feature vector;
acquiring a first anchor sample feature vector set corresponding to the anchor sample medical record data sets one by one;
inputting each first anchor sample feature vector in the first anchor sample feature vector set into a trained generator to obtain a plurality of second anchor sample feature vectors;
determining a first anchor sample average feature vector according to the plurality of second anchor sample feature vectors;
performing vector operation on the first feature vector and the first anchor sample average feature vector to obtain a second feature vector;
and inputting the second feature vector into a trained discriminator to obtain a quality detection result.
A second aspect of the present application provides a quality detection apparatus for medical record data, including:
the acquisition module is used for acquiring medical record data to be subjected to quality detection;
the processing module is used for vectorizing the medical record data to be subjected to quality detection to obtain a first characteristic vector;
the acquisition module is further used for acquiring first anchor sample feature vector sets corresponding to the anchor sample medical record data sets one by one;
the processing module is further configured to input each first anchor sample feature vector in the first anchor sample feature vector set into a trained generator to obtain a plurality of second anchor sample feature vectors;
the processing module is further configured to determine a first anchor sample average feature vector according to the plurality of second anchor sample feature vectors;
the processing module is further configured to perform vector operation on the first feature vector and the first anchor sample average feature vector to obtain a second feature vector;
and the processing module is also used for inputting the second feature vector into a trained discriminator to obtain a quality detection result.
A third aspect of the application provides an electronic device for quality testing of medical record data, comprising a processor, a memory, a communication interface, and one or more programs, wherein the one or more programs are stored in the memory and are generated as instructions to be executed by the processor to perform steps in any one of the methods for quality testing of medical record data.
A fourth aspect of the present application provides a computer-readable storage medium for storing a computer program, which is executed by the processor to implement any one of the methods for quality testing of medical record data.
It can be seen that, in the above technical scheme, medical record data to be quality-detected is obtained; vectorizing the medical record data to be subjected to quality detection to obtain a first feature vector; acquiring a first anchor sample feature vector set corresponding to the anchor sample medical record data sets one by one; inputting each first anchor sample feature vector in the first anchor sample feature vector set into a trained generator to obtain a plurality of second anchor sample feature vectors; determining a first anchor sample average feature vector according to the plurality of second anchor sample feature vectors; performing vector operation on the first feature vector and the first anchor sample average feature vector to obtain a second feature vector; and inputting the second feature vector into a trained discriminator to obtain a quality detection result. The quality detection method has the advantages that the quality detection result is obtained by inputting the second characteristic vector obtained by carrying out vector operation on the first characteristic vector corresponding to the medical record data to be quality detected and the average characteristic vector of the first anchor sample into the trained discriminator, and the problem of inaccurate quality detection result obtained by directly processing the first characteristic vector corresponding to the medical record data to be quality detected by using the trained discriminator is solved. Meanwhile, the quality detection result of the medical record data to be subjected to quality detection is determined by the discriminator, so that the quality detection efficiency of the medical record data is improved, and the quality condition of the medical record data can be reflected in time.
Drawings
In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments of the present application, and for those skilled in the art, other drawings can be obtained according to the drawings without creative efforts.
Wherein:
fig. 1 is a schematic diagram of a system for quality inspection of medical record data according to an embodiment of the present application;
fig. 2A is a schematic flowchart of a method for quality inspection of medical record data according to an embodiment of the present application;
fig. 2B is a schematic diagram of an encoder according to an embodiment of the present application;
fig. 3 is a schematic flowchart of another method for quality inspection of medical record data according to an embodiment of the present application;
fig. 4 is a schematic diagram of a device for quality inspection of medical record data according to an embodiment of the present application;
fig. 5 is a schematic structural diagram of an electronic device in a hardware operating environment according to an embodiment of the present application.
Detailed Description
The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it is obvious that the described embodiments are only a part of the embodiments of the present application, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present application.
The following are detailed below.
The terms "first" and "second" in the description and claims of the present application and the above-described drawings are used for distinguishing between different objects and not for describing a particular order. Furthermore, the terms "include" and "have," as well as any variations thereof, are intended to cover non-exclusive inclusions. For example, a process, method, system, article, or apparatus that comprises a list of steps or elements is not limited to only those steps or elements listed, but may alternatively include other steps or elements not listed, or inherent to such process, method, article, or apparatus.
Referring to fig. 1, fig. 1 is a schematic diagram of a quality detection system for medical record data according to an embodiment of the present application, where the quality detection system 100 includes a quality detection processing device 110. The quality testing processing device 110 is used for processing medical record data to be tested for quality. The quality detection system 100 may comprise an integrated single device or multiple devices, and for convenience of description, the quality detection system 100 is referred to herein as an electronic device. It will be apparent that the electronic device may include various handheld devices, vehicle-mounted devices, wearable devices, computing devices or other processing devices connected to a wireless modem having wireless communication capability, as well as various forms of User Equipment (UE), Mobile Stations (MS), terminal Equipment (terminal device), and the like.
Generally, no matter the medical records are written manually or filled by an electronic medical record system, doctors are required to check medical record data, so that the quality of the medical record data is controlled in real time. That is, in the prior art, quality control of medical record data is completed manually, and this quality detection method is inefficient and cannot reflect quality of medical record data in time.
Based on this, the embodiment of the present application provides a quality detection method for medical record data to solve the above problem, and the following describes the embodiment of the present application in detail.
Referring to fig. 2A, fig. 2A is a schematic flowchart of a method for quality detection of medical record data according to an embodiment of the present application. As shown in fig. 2A, the method includes:
201. acquiring medical record data to be subjected to quality detection;
the medical record data to be quality-detected may include characters, symbols, charts, graphs, data, images, and the like. Further, the medical record data to be quality-tested includes sex, age, year and month of birth, name of medicine, etc.
In addition, medical record data to be quality tested can be obtained from the block chain.
The block chain is a chain data structure which connects the data blocks according to the time sequence, and is a distributed account book which is cryptographically guaranteed to be not falsifiable and counterfeitable. The blockchain may include a blockchain underlying platform, a platform product service layer, an application service layer, and the like.
Further, the properties of the blockchain include openness, consensus, de-centering, de-trust, transparency, anonymity of both sides, non-tampering, traceability, and the like. Open and transparent means that anyone can participate in the blockchain network, and each device can be used as a node, and each node allows a complete database copy to be obtained. The nodes maintain the whole block chain together through competition calculation based on a set of consensus mechanism. When any node fails, the rest nodes can still work normally. The decentralization and the distrust mean that a block chain is formed into an end-to-end network by a plurality of nodes together, and no centralized equipment or management mechanism exists. The data exchange between the nodes is verified by a digital signature technology, mutual trust is not needed, and other nodes cannot be deceived as long as the data exchange is carried out according to the rules set by the system. Transparent and anonymous meaning that the operation rule of the block chain is public, and all data information is also public, so that each transaction is visible to all nodes. Because the nodes are distrusted, the nodes do not need to disclose identities, and each participated node is anonymous. Among other things, non-tamperable and traceable means that modifications to the database by each and even multiple nodes cannot affect the databases of other nodes unless more than 51% of the nodes in the entire network can be controlled to modify at the same time, which is almost impossible. In the block chain, each transaction is connected with two adjacent blocks in series through a cryptographic method, so that any transaction record can be traced.
In particular, the blockchain may utilize blockchain data structures to verify and store data, utilize distributed node consensus algorithms to generate and update data, cryptographically secure data transmission and access, and utilize intelligent contracts comprised of automated script code to program and manipulate data in a completely new distributed infrastructure and computing manner. Therefore, the characteristic that the block chain technology is not tampered fundamentally changes a centralized credit creation mode, and the irrevocability and the safety of data are effectively improved. The intelligent contract enables all the terms to be written into programs, the terms can be automatically executed on the block chain, and therefore when conditions for triggering the intelligent contract exist, the block chain can be forcibly executed according to the content in the intelligent contract and is not blocked by any external force, effectiveness and execution force of the contract are guaranteed, cost can be greatly reduced, and efficiency can be improved. Each node on the block chain has the same account book, and the recording process of the account book can be ensured to be public and transparent. The block chain technology can realize point-to-point, open and transparent direct interaction, so that an information interaction mode with high efficiency, large scale and no centralized agent becomes a reality.
202. Vectorizing the medical record data to be subjected to quality detection to obtain a first feature vector;
it should be noted that the data type corresponding to the medical record data to be quality-detected includes a continuous type or a classification type, and before vectorizing the medical record data to be quality-detected to obtain the first feature vector, the method further includes: and coding the data with the data type being the classification type in the medical record data to be quality detected to obtain the coded medical record data to be quality detected. The medical record data to be quality-detected can be obtained by encoding data of which the data type is the type in the medical record data to be quality-detected by adopting one-hot encoding. It can be understood that the data of which the data type is continuous in the medical record data to be quality-detected does not need to be encoded.
For example, if the data type corresponding to the gender in the medical record data to be quality-checked is a type, encoding the gender is required. And the data type corresponding to the age in the medical record data to be subjected to quality detection is continuous, so that the age does not need to be coded.
Further, vectorizing the medical record data to be subjected to quality detection to obtain a first feature vector includes: vectorizing the encoded medical record data to be subjected to quality detection and the data of which the data type is continuous in the medical record data to be subjected to quality detection to obtain a first feature vector. It can be understood that the encoded medical record data to be quality-detected and the data of which the data type is continuous in the medical record data to be quality-detected may be input into the encoder to obtain the first feature vector.
Referring to fig. 2B, fig. 2B is a schematic diagram of an encoder according to an embodiment of the present disclosure. As shown in fig. 2B, it can be seen that the encoder includes an input layer, at least one hidden layer, and an output layer. It should be noted that the input layer is an n-dimensional input layer, the hidden layer is an m-dimensional hidden layer, and the output layer is an n-dimensional output layer. Wherein n and m are integers greater than 1, and m is much less than n. Furthermore, the mapping of the hidden layer is used as an encoder, the mapping of the output layer is used as a decoder, and the network structure of the encoder is a multi-layer self-coding network structure.
It should be noted that the encoded medical record data to be quality-detected and the data of which the data type is continuous in the medical record data to be quality-detected are n-dimensional high-dimensional vectors. The first feature vector is a low-dimensional vector of m dimensions, i.e. the first feature vector is the output data of the last hidden layer.
203. Acquiring a first anchor sample feature vector set corresponding to the anchor sample medical record data sets one by one;
optionally, in a possible implementation manner, the acquiring a first anchor sample feature vector set corresponding to the anchor sample medical record data sets one to one includes: acquiring a sample medical record data set to be trained and a quality score corresponding to each sample medical record data to be trained in the sample medical record data set to be trained; determining sample medical record data to be trained with quality scores in a first preset score interval according to the quality scores corresponding to the sample medical record data to be trained in the sample medical record data set to be trained, and obtaining an anchor sample medical record data set; vectorizing each piece of anchor sample medical record data in the anchor sample medical record data set to obtain the first anchor sample feature vector set.
The medical record data of the sample to be trained with the quality score in a second preset score interval can be determined according to the quality score corresponding to the medical record data of each sample to be trained in the medical record data set of the sample to be trained, so that a positive sample medical record data set is obtained; and determining the sample medical record data to be trained with the quality score in a third preset score interval according to the quality score corresponding to each sample medical record data to be trained in the sample medical record data set to be trained, so as to obtain a negative sample medical record data set. It can be understood that the first preset scoring interval is higher than the second preset scoring interval, and the second preset scoring interval is higher than the third preset scoring interval. That is, the quality score corresponding to each anchor sample medical record data in the anchor sample medical record data set is higher than the quality score corresponding to each positive sample medical record data in the positive sample medical record data set, and the quality score corresponding to each positive sample medical record data in the positive sample medical record data set is higher than the quality score corresponding to each negative sample medical record data in the negative sample medical record data set. In addition, each piece of negative sample medical record data in the negative sample medical record data set has a quality problem. However, each piece of positive sample medical record data in the positive sample medical record data set and each piece of anchor sample medical record data in the anchor sample medical record data set have no quality problem.
For example, the completion time, writing format paragraphs, medical terminology, third-level ward visit, informed consent, anesthesia visit, diagnosis and treatment, auxiliary examination, nosocomial infection and/or antibacterial drug use of the medical record of the sample medical record data with the quality score in the first preset scoring interval meet the regulations; medical terms, third-level ward visit, informed consent, anesthesia visit, diagnosis and treatment, auxiliary examination, nosocomial infection and/or antibacterial drug use of the medical record data of the sample to be trained with the quality score in the second preset scoring interval meet the regulations; medical terms, third-level ward round, anesthesia visit, diagnosis and treatment, nosocomial infection and/or antibacterial drug use of the medical record data of the sample to be trained with the quality score in the third preset scoring interval meet the regulations.
It should be noted that the data type corresponding to each sample medical record data to be trained in the sample medical record data set to be trained includes a continuous type or a classification type. That is, the data type corresponding to each anchor sample medical record data in the anchor sample medical record data set includes a continuous type or a category type, the data type corresponding to each positive sample medical record data in the positive sample medical record data set includes a continuous type or a category type, and the data type corresponding to each negative sample medical record data in the negative sample medical record data set includes a continuous type or a category type. Further, the anchor sample medical record data P is any one of the anchor sample medical record data sets, and before vectorizing each of the anchor sample medical record data in the anchor sample medical record data set to obtain the first anchor sample feature vector set, the method further includes: and coding the data with the data type of the type in the anchor sample medical record data P to obtain the coded anchor sample medical record data P. The anchor sample medical record data P can be encoded by using unique hot coding, so as to obtain encoded anchor sample medical record data P. It can be understood that the data with the data type being continuous in the anchor sample medical record data P does not need to be encoded.
Further, vectorizing each anchor sample medical record data in the anchor sample medical record data set to obtain the first anchor sample feature vector set, including: vectorizing the encoded anchor sample medical record data P and data of which the data type is continuous in the anchor sample medical record data P to obtain a first anchor sample feature vector corresponding to the anchor sample medical record data P. It can be understood that the encoded anchor sample medical record data P and the data of the anchor sample medical record data P with the continuous data type may be input into the encoder to obtain the first anchor sample feature vector corresponding to the anchor sample medical record data P. As can be appreciated, the anchor sample medical record data set corresponds one-to-one to the first anchor sample feature vector set.
In addition, the encoded anchor sample medical record data P and the data of which the data type is continuous in the anchor sample medical record data P are n-dimensional high-dimensional vectors. The first anchor sample feature vector corresponding to the anchor sample medical record data P is a m-dimensional low-dimensional vector, that is, the first anchor sample feature vector corresponding to the anchor sample medical record data P is output data of the last hidden layer.
It can be seen that, in the above technical solution, a sample medical record data set to be trained and a quality score corresponding to each sample medical record data to be trained in the sample medical record data set to be trained are obtained; determining sample medical record data to be trained with quality scores in a first preset score interval according to the quality scores corresponding to the sample medical record data to be trained in the sample medical record data set to be trained, and obtaining an anchor sample medical record data set; vectorizing each piece of anchor sample medical record data in the anchor sample medical record data set to obtain the first anchor sample feature vector set. The classification of the medical record data sets of the samples to be trained is realized by determining the medical record data of the samples to be trained with the quality scores in the first preset scoring interval, and the medical record data sets of the anchor samples with the quality scores in the first preset scoring interval are also obtained. Meanwhile, vectorizing each piece of anchor sample medical record data in the anchor sample medical record data set to obtain an anchor sample feature vector set, and preparing for subsequently determining an anchor sample average feature vector. In addition, the high-dimensional vector is converted into the low-dimensional vector through vectorization, so that the distribution difficulty of the trained generator for learning the medical record data of the anchor sample is simplified, and the learning efficiency is improved.
204. Inputting each first anchor sample feature vector in the first anchor sample feature vector set into a trained generator to obtain a plurality of second anchor sample feature vectors;
205. determining a first anchor sample average feature vector according to the plurality of second anchor sample feature vectors;
wherein the first anchor sample average feature vector is a low-dimensional vector of m dimensions. The ith row and jth column value in the first anchor sample average feature vector is an average of the ith row and jth column values in each of the plurality of second anchor sample feature vectors. Wherein i and j are integers greater than 0, and the values of i and j are related to the first anchor sample average feature vector.
For example, the plurality of second anchor sample feature vectors includes second anchor sample feature vector N1 and second anchor sample feature vector N2. Wherein the second anchor sample feature vector N1 is
The second anchor sample feature vector N2 is
Then, the first anchor sample average feature vector is
206. Performing vector operation on the first feature vector and the first anchor sample average feature vector to obtain a second feature vector;
the vector operation may be, for example, vector addition, vector subtraction, vector product, etc., and is not limited herein.
207. And inputting the second feature vector into a trained discriminator to obtain a quality detection result.
It can be seen that, in the above technical scheme, medical record data to be quality-detected is obtained; vectorizing the medical record data to be subjected to quality detection to obtain a first feature vector; acquiring a first anchor sample feature vector set corresponding to the anchor sample medical record data sets one by one; inputting each first anchor sample feature vector in the first anchor sample feature vector set into a trained generator to obtain a plurality of second anchor sample feature vectors; determining a first anchor sample average feature vector according to the plurality of second anchor sample feature vectors; performing vector operation on the first feature vector and the first anchor sample average feature vector to obtain a second feature vector; and inputting the second feature vector into a trained discriminator to obtain a quality detection result. The quality detection method has the advantages that the quality detection result is obtained by inputting the second characteristic vector obtained by carrying out vector operation on the first characteristic vector corresponding to the medical record data to be quality detected and the average characteristic vector of the first anchor sample into the trained discriminator, and the problem of inaccurate quality detection result obtained by directly processing the first characteristic vector corresponding to the medical record data to be quality detected by using the trained discriminator is solved. Meanwhile, the quality detection result of the medical record data to be subjected to quality detection is determined by the discriminator, so that the quality detection efficiency of the medical record data is improved, and the quality condition of the medical record data can be reflected in time.
In a possible implementation manner, the inputting the second feature vector into a trained discriminator to obtain a quality detection result includes: inputting the second feature vector into a trained discriminator to obtain a quality detection value; when the quality detection value is higher than a threshold value, determining that the quality detection result is that the medical record data to be quality detected has no quality problem; and when the quality detection value is lower than a threshold value, determining that the quality detection result is that the medical record data to be subjected to quality detection has a quality problem.
The quality detection numerical value is a floating point number in a preset interval, and the preset interval is [0,1 ]. Further, when the quality detection value is higher than the threshold, the label is 1, that is, the quality detection result is that the medical record data to be quality-detected has no quality problem; when the quality detection value is lower than the threshold value, the label is 0, and the quality detection result indicates that the medical record data to be subjected to quality detection has quality problems.
Wherein, a threshold adjustment interface can also be displayed, the threshold adjustment interface comprising a threshold input box and a confirmation button. The user can input the threshold value in the threshold value input box and operate the confirmation button, so that the dynamic adjustment of the threshold value is realized.
It can be seen that, in the above technical solution, the second feature vector is input into a trained discriminator to obtain a quality detection value; when the quality detection value is higher than a threshold value, determining that the quality detection result is that the medical record data to be quality detected has no quality problem; when the quality detection value is lower than the threshold value, the quality detection result is determined to be that the medical record data to be quality detected has a quality problem, and the quality detection result of the medical record data to be quality detected is determined by utilizing the discriminator, so that the quality detection efficiency of the medical record data is improved, and the quality condition of the medical record data can be reflected in time. Meanwhile, the quality detection result of the medical record data is dynamically controlled by combining the threshold value.
Referring to fig. 3, fig. 3 is a schematic flowchart of another method for quality detection of medical record data according to an embodiment of the present application. Wherein, as shown in fig. 3, the method further comprises:
301. inputting each first anchor sample feature vector in the first anchor sample feature vector set into a generator to be trained to obtain a plurality of third anchor sample feature vectors;
wherein the generator to be trained comprises an input layer, a plurality of hidden layers and an output layer. It should be noted that the input layer is an m-dimensional input layer, the hidden layer is a k-dimensional hidden layer, and the output layer is an m-dimensional output layer. Wherein k is an integer greater than 1 and less than m. Further, the network structure of the generator to be trained is a deep neural network.
In addition, each first anchor sample feature vector in the first anchor sample feature vector set is a low-dimensional vector of m dimensions. Each of the plurality of third anchor sample feature vectors is a low-dimensional vector of m-dimensions, i.e., each of the plurality of third anchor sample feature vectors is output data of an output layer of the generator to be trained.
It will be appreciated that the generators to be trained and the generators trained differ significantly in their internal parameters. Therefore, when each first anchor sample feature vector in the first anchor sample feature vector set is respectively input into the generator to be trained and the generator which is trained, the output data output by the generator to be trained is greatly different from the output data output by the generator which is trained.
In addition, each first anchor sample feature vector in the first anchor sample feature vector set is input into a generator to be trained, so that training of the generator can be completed, and the trained generator is obtained.
302. Determining a second anchor sample average feature vector according to the plurality of third anchor sample feature vectors;
wherein the second anchor sample average feature vector is a m-dimensional low-dimensional vector. The row a and column b values in the second anchor sample average feature vector are the average of the row a and column b values in each of the plurality of third anchor sample feature vectors. And a and b are integers larger than 0, and the values of a and b are related to the average feature vector of the second anchor sample.
For example, the plurality of third anchor sample feature vectors includes third anchor sample feature vector M1 and third anchor sample feature vector M2. Wherein the third anchor sample feature vector M1 is
The third anchor sample feature vector M2 is
Then the second anchor sample average feature vector is
303. Acquiring a positive sample feature vector set and a negative sample feature vector set corresponding to the positive sample medical record data sets one to one;
optionally, in a possible implementation manner, the acquiring of the positive sample feature vector set corresponding to the positive sample medical record data set includes: encoding the data with the data type of the classification in the positive sample medical record data X to obtain encoded positive sample medical record data X; vectorizing the coded positive sample medical record data X and the data of which the data type is continuous in the positive sample medical record data X to obtain a positive sample feature vector corresponding to the positive sample medical record data X.
The data with the data type of the positive sample medical record data X being the type can be encoded by using the one-hot encoding, so as to obtain the encoded positive sample medical record data X. It can be understood that the data of the data type continuous type in the positive sample medical record data X need not be encoded.
In addition, the encoded positive sample medical record data X and the data of which the data type is continuous in the positive sample medical record data X are n-dimensional high-dimensional vectors. The positive sample feature vector corresponding to the positive sample medical record data X is a m-dimensional low-dimensional vector, that is, the positive sample feature vector corresponding to the positive sample medical record data X is output data of the last hidden layer.
Similarly, the negative sample medical record data Y is any piece of data in the negative sample medical record data set, and the obtaining of the negative sample feature vector set corresponding to the negative sample medical record data set one to one includes: encoding the data with the data type of the category in the negative sample medical record data Y to obtain encoded negative sample medical record data Y; vectorizing the encoded negative sample medical record data Y and the data of which the data type is continuous in the negative sample medical record data Y to obtain a negative sample feature vector corresponding to the negative sample medical record data Y.
The negative sample medical record data Y can be encoded by using a single hot code, so as to obtain encoded negative sample medical record data Y. It can be understood that the data with the continuous data type in the negative sample medical record data Y does not need to be encoded.
In addition, the encoded negative sample medical record data Y and the data of which the data type is continuous in the negative sample medical record data Y are n-dimensional high-dimensional vectors. The negative sample feature vector corresponding to the negative sample medical record data Y is a m-dimensional low-dimensional vector, that is, the negative sample feature vector corresponding to the negative sample medical record data Y is output data of the last hidden layer.
304. Performing vector operation on the second anchor sample average feature vector and each positive sample feature vector in the positive sample feature vector set to obtain a first sample feature vector set;
the first sample feature vector a is any one of the first sample feature vector set, a value of the first sample feature vector a is used to represent a distance between a positive sample feature vector corresponding to the first sample feature vector a and the second anchor sample average feature vector, when the value of the first sample feature vector a is larger, the positive sample feature vector corresponding to the first sample feature vector a is closer to the second anchor sample average feature vector, and when the value of the first sample feature vector a is smaller, the positive sample feature vector corresponding to the first sample feature vector a is farther from the second anchor sample average feature vector.
305. Performing vector operation on the second anchor sample average feature vector and each negative sample feature vector in the negative sample feature vector set to obtain a second sample feature vector set;
the second sample feature vector B is any one of the second sample feature vector set, a value of the second sample feature vector B is used to represent a distance between a negative sample feature vector corresponding to the second sample feature vector B and the second anchor sample average feature vector, when a value of the second sample feature vector B is larger, the negative sample feature vector corresponding to the second sample feature vector B is closer to the second anchor sample average feature vector, and when the value of the second sample feature vector B is smaller, the negative sample feature vector corresponding to the second sample feature vector B is farther from the second anchor sample average feature vector.
306. And respectively inputting the first sample feature vector set and the second sample feature vector set into a discriminator to be trained to obtain the trained discriminator.
It can be seen that, in the above technical solution, each first anchor sample feature vector in the first anchor sample feature vector set is input into a generator to be trained to obtain a plurality of third anchor sample feature vectors; determining a second anchor sample average feature vector according to the plurality of third anchor sample feature vectors; acquiring a positive sample feature vector set and a negative sample feature vector set corresponding to the positive sample medical record data sets one to one; performing vector operation on the second anchor sample average feature vector and each positive sample feature vector in the positive sample feature vector set to obtain a first sample feature vector set; performing vector operation on the second anchor sample average feature vector and each negative sample feature vector in the negative sample feature vector set to obtain a second sample feature vector set; and respectively inputting the first sample characteristic vector set and the second sample characteristic vector set into a discriminator to be trained to obtain the trained discriminator, obtaining an anchor sample average characteristic vector by using the anchor sample characteristic vector generated by the generator to be trained, and training the discriminator to be trained by using the vector obtained after vector operation is respectively carried out on the anchor sample average characteristic vector, the positive sample characteristic vector and the negative sample characteristic vector, so that the farther the negative sample and the anchor sample are away from each other, the closer the positive sample and the anchor sample are to each other, and the trained discriminator can better classify medical record data with quality problems and medical record data without quality problems. Simultaneously, this scheme can be applied to in the wisdom medical field, can better classify out the case history data that has the quality problem and the case history data that does not have the quality problem through letting the arbiter that trains well to the construction in wisdom city has been promoted better.
Referring to fig. 4, fig. 4 is a schematic diagram of a quality detection apparatus for medical record data according to an embodiment of the present application. As shown in fig. 4, an apparatus 400 for quality detection of medical record data according to an embodiment of the present application may include:
an obtaining module 401, configured to obtain medical record data to be quality-detected;
the processing module 402 is configured to perform vectorization on the medical record data to be quality-detected to obtain a first feature vector;
the obtaining module 401 is further configured to obtain a first anchor sample feature vector set corresponding to the anchor sample medical record data sets one to one;
optionally, in a possible implementation manner, when acquiring a first anchor sample feature vector set corresponding to an anchor sample medical record data set one to one, the acquiring module 401 is specifically configured to acquire a sample medical record data set to be trained and a quality score corresponding to each sample medical record data to be trained in the sample medical record data set to be trained; the processing module 402 is specifically configured to determine, according to a quality score corresponding to each sample medical record data to be trained in the sample medical record data set to be trained, sample medical record data to be trained, of which the quality score is in a first preset score interval, and obtain an anchor sample medical record data set; vectorizing each piece of anchor sample medical record data in the anchor sample medical record data set to obtain the first anchor sample feature vector set.
The processing module 402 is further configured to input each first anchor sample feature vector in the first anchor sample feature vector set into a trained generator, so as to obtain a plurality of second anchor sample feature vectors;
the processing module 402 is further configured to determine a first anchor sample average feature vector according to the plurality of second anchor sample feature vectors;
the processing module 402 is further configured to perform vector operation on the first feature vector and the first anchor sample average feature vector to obtain a second feature vector;
the processing module 402 is further configured to input the second feature vector into a trained discriminator to obtain a quality detection result.
Optionally, in a possible implementation manner, the processing module 402 is further configured to input each first anchor sample feature vector in the first anchor sample feature vector set into a generator to be trained, so as to obtain a plurality of third anchor sample feature vectors; the processing module 402 is further configured to determine a second anchor sample average feature vector according to the plurality of third anchor sample feature vectors; the obtaining module 401 is further configured to obtain a positive sample feature vector set corresponding to the positive sample medical record data set one to one and a negative sample feature vector set corresponding to the negative sample medical record data set one to one; the processing module 402 is further configured to perform vector operation on the second anchor sample average feature vector and each positive sample feature vector in the positive sample feature vector set to obtain a first sample feature vector set; the processing module 402 is further configured to perform vector operation on the second anchor sample average feature vector and each negative sample feature vector in the negative sample feature vector set to obtain a second sample feature vector set; the processing module 402 is further configured to input the first sample feature vector set and the second sample feature vector set into a to-be-trained discriminator, respectively, to obtain the trained discriminator.
The first sample feature vector a is any one of the first sample feature vector set, a value of the first sample feature vector a is used to represent a distance between a positive sample feature vector corresponding to the first sample feature vector a and the second anchor sample average feature vector, when the value of the first sample feature vector a is larger, the positive sample feature vector corresponding to the first sample feature vector a is closer to the second anchor sample average feature vector, and when the value of the first sample feature vector a is smaller, the positive sample feature vector corresponding to the first sample feature vector a is farther from the second anchor sample average feature vector.
The second sample feature vector B is any one of the second sample feature vector set, a value of the second sample feature vector B is used to represent a distance between a negative sample feature vector corresponding to the second sample feature vector B and the second anchor sample average feature vector, when a value of the second sample feature vector B is larger, the negative sample feature vector corresponding to the second sample feature vector B is closer to the second anchor sample average feature vector, and when the value of the second sample feature vector B is smaller, the negative sample feature vector corresponding to the second sample feature vector B is farther from the second anchor sample average feature vector.
Optionally, in a possible implementation manner, when the second feature vector is input into a trained discriminator to obtain a quality detection result, the processing module 402 is specifically configured to input the second feature vector into the trained discriminator to obtain a quality detection value; when the quality detection value is higher than a threshold value, determining that the quality detection result is that the medical record data to be quality detected has no quality problem; and when the quality detection value is lower than a threshold value, determining that the quality detection result is that the medical record data to be subjected to quality detection has a quality problem.
Referring to fig. 5, fig. 5 is a schematic structural diagram of an electronic device in a hardware operating environment according to an embodiment of the present application.
The embodiment of the application provides an electronic device for quality detection of medical record data, which comprises a processor, a memory, a communication interface and one or more programs, wherein the one or more programs are stored in the memory and are configured to be executed by the processor to execute instructions of steps in a quality detection method of any medical record data. As shown in fig. 5, an electronic device of a hardware operating environment according to an embodiment of the present application may include:
a processor 501, such as a CPU.
The memory 502 may alternatively be a high speed RAM memory or a stable memory such as a disk memory.
A communication interface 503 for implementing connection communication between the processor 501 and the memory 502.
Those skilled in the art will appreciate that the configuration of the electronic device shown in fig. 5 is not intended to be limiting and may include more or fewer components than shown, or some components in combination, or a different arrangement of components.
As shown in fig. 5, the memory 502 may include an operating system, a network communication module, and a program for verification of gray scale distribution. An operating system is a program that manages and controls the server hardware and software resources, supporting the execution of one or more programs. The network communication module is used for communication among the components in the memory 502 and with other hardware and software in the electronic device.
In the electronic device shown in fig. 5, the processor 501 is configured to execute a program for personnel management stored in the memory 502, and implement the following steps: acquiring medical record data to be subjected to quality detection; vectorizing the medical record data to be subjected to quality detection to obtain a first feature vector; acquiring a first anchor sample feature vector set corresponding to the anchor sample medical record data sets one by one; inputting each first anchor sample feature vector in the first anchor sample feature vector set into a trained generator to obtain a plurality of second anchor sample feature vectors; determining a first anchor sample average feature vector according to the plurality of second anchor sample feature vectors; performing vector operation on the first feature vector and the first anchor sample average feature vector to obtain a second feature vector; and inputting the second feature vector into a trained discriminator to obtain a quality detection result.
For specific implementation of the electronic device related to the present application, reference may be made to various embodiments of the quality detection method for medical record data, which are not described herein again.
The present application further provides a computer readable storage medium for storing a computer program, the stored computer program being executable by the processor to perform the steps of: acquiring medical record data to be subjected to quality detection; vectorizing the medical record data to be subjected to quality detection to obtain a first feature vector; acquiring a first anchor sample feature vector set corresponding to the anchor sample medical record data sets one by one; inputting each first anchor sample feature vector in the first anchor sample feature vector set into a trained generator to obtain a plurality of second anchor sample feature vectors; determining a first anchor sample average feature vector according to the plurality of second anchor sample feature vectors; performing vector operation on the first feature vector and the first anchor sample average feature vector to obtain a second feature vector; and inputting the second feature vector into a trained discriminator to obtain a quality detection result.
For specific implementation of the computer-readable storage medium according to the present application, reference may be made to the embodiments of the medical record data quality detection method, which are not described herein again.
It should be noted that, for simplicity of description, the above-mentioned method embodiments are described as a series of acts or combination of acts, but those skilled in the art should understand that the present application is not limited by the order of acts described, as some steps may be performed in other orders or simultaneously according to the present application. Further, those skilled in the art should also appreciate that the embodiments described in this specification are preferred embodiments and that the acts and modules involved are not necessarily required for this application.
The above embodiments are only used for illustrating the technical solutions of the present application, and not for limiting the same; although the present application has been described in detail with reference to the foregoing embodiments, it should be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; and the modifications or the substitutions do not make the essence of the corresponding technical solutions depart from the scope of the technical solutions of the embodiments of the present application.
Claims (10)
1. A method for detecting the quality of medical record data is characterized by comprising the following steps:
acquiring medical record data to be subjected to quality detection;
vectorizing the medical record data to be subjected to quality detection to obtain a first feature vector;
acquiring a first anchor sample feature vector set corresponding to the anchor sample medical record data sets one by one;
inputting each first anchor sample feature vector in the first anchor sample feature vector set into a trained generator to obtain a plurality of second anchor sample feature vectors;
determining a first anchor sample average feature vector according to the plurality of second anchor sample feature vectors;
performing vector operation on the first feature vector and the first anchor sample average feature vector to obtain a second feature vector;
and inputting the second feature vector into a trained discriminator to obtain a quality detection result.
2. The method of claim 1, wherein obtaining a first set of anchor sample feature vectors in a one-to-one correspondence with a set of anchor sample medical record data comprises:
acquiring a sample medical record data set to be trained and a quality score corresponding to each sample medical record data to be trained in the sample medical record data set to be trained;
determining sample medical record data to be trained with quality scores in a first preset score interval according to the quality scores corresponding to the sample medical record data to be trained in the sample medical record data set to be trained, and obtaining an anchor sample medical record data set;
vectorizing each piece of anchor sample medical record data in the anchor sample medical record data set to obtain the first anchor sample feature vector set.
3. The method according to claim 1 or 2, characterized in that the method further comprises:
inputting each first anchor sample feature vector in the first anchor sample feature vector set into a generator to be trained to obtain a plurality of third anchor sample feature vectors;
determining a second anchor sample average feature vector according to the plurality of third anchor sample feature vectors;
acquiring a positive sample feature vector set and a negative sample feature vector set corresponding to the positive sample medical record data sets one to one;
performing vector operation on the second anchor sample average feature vector and each positive sample feature vector in the positive sample feature vector set to obtain a first sample feature vector set;
performing vector operation on the second anchor sample average feature vector and each negative sample feature vector in the negative sample feature vector set to obtain a second sample feature vector set;
and respectively inputting the first sample feature vector set and the second sample feature vector set into a discriminator to be trained to obtain the trained discriminator.
4. The method according to claim 3, wherein a first sample feature vector A is any one of the first set of sample feature vectors, a value of the first sample feature vector A is used to indicate a distance between a positive sample feature vector corresponding to the first sample feature vector A and the second anchor sample average feature vector, the positive sample feature vector corresponding to the first sample feature vector A is closer to the second anchor sample average feature vector when the value of the first sample feature vector A is larger, and the positive sample feature vector corresponding to the first sample feature vector A is farther from the second anchor sample average feature vector when the value of the first sample feature vector A is smaller.
5. The method according to claim 3, wherein a second sample feature vector B is any one of the second sample feature vector set, and a value of the second sample feature vector B is used to indicate a distance between a negative sample feature vector corresponding to the second sample feature vector B and the second anchor sample average feature vector, and the negative sample feature vector corresponding to the second sample feature vector B is closer to the second anchor sample average feature vector when the value of the second sample feature vector B is larger, and the negative sample feature vector corresponding to the second sample feature vector B is farther from the second anchor sample average feature vector when the value of the second sample feature vector B is smaller.
6. The method of claim 1, wherein inputting the second feature vector into a trained discriminator to obtain a quality detection result comprises:
inputting the second feature vector into a trained discriminator to obtain a quality detection value;
when the quality detection value is higher than a threshold value, determining that the quality detection result is that the medical record data to be quality detected has no quality problem;
and when the quality detection value is lower than a threshold value, determining that the quality detection result is that the medical record data to be subjected to quality detection has a quality problem.
7. A quality detection device for medical record data is characterized by comprising:
the acquisition module is used for acquiring medical record data to be subjected to quality detection;
the processing module is used for vectorizing the medical record data to be subjected to quality detection to obtain a first characteristic vector;
the acquisition module is further used for acquiring first anchor sample feature vector sets corresponding to the anchor sample medical record data sets one by one;
the processing module is further configured to input each first anchor sample feature vector in the first anchor sample feature vector set into a trained generator to obtain a plurality of second anchor sample feature vectors;
the processing module is further configured to determine a first anchor sample average feature vector according to the plurality of second anchor sample feature vectors;
the processing module is further configured to perform vector operation on the first feature vector and the first anchor sample average feature vector to obtain a second feature vector;
and the processing module is also used for inputting the second feature vector into a trained discriminator to obtain a quality detection result.
8. The apparatus of claim 7, wherein, when obtaining a first set of anchor sample feature vectors for a one-to-one correspondence of anchor sample medical record data sets,
the acquisition module is specifically used for acquiring a sample medical record data set to be trained and a quality score corresponding to each sample medical record data to be trained in the sample medical record data set to be trained;
the processing module is specifically configured to determine sample medical record data to be trained with a quality score in a first preset score interval according to the quality score corresponding to each sample medical record data to be trained in the sample medical record data set to be trained, and obtain an anchor sample medical record data set; vectorizing each piece of anchor sample medical record data in the anchor sample medical record data set to obtain the first anchor sample feature vector set.
9. An electronic device for quality testing of medical record data, comprising a processor, a memory, a communication interface, and one or more programs, wherein the one or more programs are stored in the memory and generated as instructions for execution by the processor to perform the steps of the method of any of claims 1-6.
10. A computer-readable storage medium, characterized in that the computer-readable storage medium is used to store a computer program, which is executed by the processor, to implement the method of any of claims 1-6.
Priority Applications (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010416797.9A CN111696637A (en) | 2020-05-15 | 2020-05-15 | Quality detection method and related device for medical record data |
| PCT/CN2020/099270 WO2021114626A1 (en) | 2020-05-15 | 2020-06-30 | Method for detecting quality of medical record data and related device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010416797.9A CN111696637A (en) | 2020-05-15 | 2020-05-15 | Quality detection method and related device for medical record data |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111696637A true CN111696637A (en) | 2020-09-22 |
Family
ID=72477882
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010416797.9A Pending CN111696637A (en) | 2020-05-15 | 2020-05-15 | Quality detection method and related device for medical record data |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN111696637A (en) |
| WO (1) | WO2021114626A1 (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111883222A (en) * | 2020-09-28 | 2020-11-03 | 平安科技（深圳）有限公司 | Text data error detection method and device, terminal equipment and storage medium |
Family Cites Families (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109858046B (en) * | 2018-02-09 | 2024-03-08 | 谷歌有限责任公司 | Learning long-term dependencies in neural networks using assistance loss |
| CN109003678B (en) * | 2018-06-12 | 2021-04-30 | 清华大学 | Method and system for generating simulated text medical record |
| CN109656878B (en) * | 2018-12-12 | 2020-11-06 | 中电健康云科技有限公司 | Health record data generation method and device |
| CN110335653B (en) * | 2019-06-30 | 2022-05-24 | 浙江大学 | Non-standard medical record analysis method based on openEHR medical record format |
-
2020
- 2020-05-15 CN CN202010416797.9A patent/CN111696637A/en active Pending
- 2020-06-30 WO PCT/CN2020/099270 patent/WO2021114626A1/en active Application Filing
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111883222A (en) * | 2020-09-28 | 2020-11-03 | 平安科技（深圳）有限公司 | Text data error detection method and device, terminal equipment and storage medium |
| CN111883222B (en) * | 2020-09-28 | 2020-12-22 | 平安科技（深圳）有限公司 | Text data error detection method and device, terminal equipment and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| WO2021114626A1 (en) | 2021-06-17 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112949786A (en) | Data classification identification method, device, equipment and readable storage medium | |
| CN112860841B (en) | Text emotion analysis method, device, equipment and storage medium | |
| CN111401700A (en) | Data analysis method, device, computer system and readable storage medium | |
| CN112988963B (en) | User intention prediction method, device, equipment and medium based on multi-flow nodes | |
| CN111785384A (en) | Abnormal data identification method based on artificial intelligence and related equipment | |
| CN112257578A (en) | Face key point detection method and device, electronic equipment and storage medium | |
| CN115237802A (en) | Artificial intelligence based simulation test method and related equipment | |
| WO2023109631A1 (en) | Data processing method and apparatus, device, storage medium, and program product | |
| CN113807728A (en) | Performance assessment method, device, equipment and storage medium based on neural network | |
| CN115130711A (en) | Data processing method and device, computer and readable storage medium | |
| CN111639706A (en) | Personal risk portrait generation method based on image set and related equipment | |
| CN115222443A (en) | Client group division method, device, equipment and storage medium | |
| CN116340793A (en) | Data processing method, device, equipment and readable storage medium | |
| Choi et al. | Detecting and analyzing politically-themed stocks using text mining techniques and transfer entropy—focus on the Republic of Korea’s case | |
| CN113705749A (en) | Two-dimensional code identification method, device and equipment based on deep learning and storage medium | |
| CN111696637A (en) | Quality detection method and related device for medical record data | |
| CN117235633A (en) | Mechanism classification method, mechanism classification device, computer equipment and storage medium | |
| CN116311313A (en) | Medical record report form detection method, device, equipment and medium based on artificial intelligence | |
| CN113343970B (en) | Text image detection method, device, equipment and storage medium | |
| CN116166999A (en) | Abnormal transaction data identification method, device, computer equipment and storage medium | |
| CN114625960A (en) | On-line evaluation method and device, electronic equipment and storage medium | |
| CN113850260A (en) | Key information extraction method and device, electronic equipment and readable storage medium | |
| CN114612246A (en) | Object set identification method and device, computer equipment and storage medium | |
| CN114429822A (en) | Medical record quality inspection method and device and storage medium | |
| CN113591881A (en) | Intention recognition method and device based on model fusion, electronic equipment and 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 |