CN117995346A - Medical record quality control optimization method and device, electronic equipment and storage medium - Google Patents
Medical record quality control optimization method and device, electronic equipment and storage medium Download PDFInfo
- Publication number
- CN117995346A CN117995346A CN202410406561.5A CN202410406561A CN117995346A CN 117995346 A CN117995346 A CN 117995346A CN 202410406561 A CN202410406561 A CN 202410406561A CN 117995346 A CN117995346 A CN 117995346A
- Authority
- CN
- China
- Prior art keywords
- data
- defect
- model
- quality control
- 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.)
- Granted
Links
- 238000003908 quality control method Methods 0.000 title claims abstract description 190
- 238000000034 method Methods 0.000 title claims abstract description 57
- 238000005457 optimization Methods 0.000 title claims abstract description 44
- 230000007547 defect Effects 0.000 claims abstract description 259
- 238000012549 training Methods 0.000 claims abstract description 63
- 238000010276 construction Methods 0.000 claims description 26
- 238000004590 computer program Methods 0.000 claims description 19
- 238000012360 testing method Methods 0.000 claims description 10
- 238000006243 chemical reaction Methods 0.000 claims description 6
- 238000001914 filtration Methods 0.000 claims description 6
- 230000007847 structural defect Effects 0.000 abstract description 8
- 238000010586 diagram Methods 0.000 description 7
- 230000008569 process Effects 0.000 description 6
- 230000000694 effects Effects 0.000 description 4
- 230000004044 response Effects 0.000 description 3
- 230000036541 health Effects 0.000 description 2
- 230000006872 improvement Effects 0.000 description 2
- 238000012552 review Methods 0.000 description 2
- 230000001360 synchronised effect Effects 0.000 description 2
- 238000011156 evaluation Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 238000013441 quality evaluation Methods 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
- 208000024891 symptom Diseases 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
- G16H15/00—ICT specially adapted for medical reports, e.g. generation or transmission thereof
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N5/00—Computing arrangements using knowledge-based models
- G06N5/02—Knowledge representation; Symbolic representation
- G06N5/022—Knowledge engineering; Knowledge acquisition
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H50/00—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics
- G16H50/70—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for mining of medical data, e.g. analysing previous cases of other patients
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
- Y02T10/00—Road transport of goods or passengers
- Y02T10/10—Internal combustion engine [ICE] based vehicles
- Y02T10/40—Engine management systems
Landscapes
- Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Public Health (AREA)
- Medical Informatics (AREA)
- Health & Medical Sciences (AREA)
- Theoretical Computer Science (AREA)
- General Engineering & Computer Science (AREA)
- Primary Health Care (AREA)
- General Health & Medical Sciences (AREA)
- Epidemiology (AREA)
- Evolutionary Computation (AREA)
- Software Systems (AREA)
- Mathematical Physics (AREA)
- Biomedical Technology (AREA)
- Databases & Information Systems (AREA)
- General Physics & Mathematics (AREA)
- Pathology (AREA)
- Computing Systems (AREA)
- Physics & Mathematics (AREA)
- Computational Linguistics (AREA)
- Artificial Intelligence (AREA)
- Medical Treatment And Welfare Office Work (AREA)
Abstract
The invention relates to a medical record quality control optimization method, a medical record quality control optimization device, electronic equipment and a storage medium, wherein the method comprises the following steps: and acquiring medical record quality control rules and first text data, wherein the medical record quality control rules comprise medical record quality control ranges and medical record quality control definitions, and the first text data is text data to be quality controlled. And respectively converting the medical record quality control rule into a first instruction and a second instruction, wherein the first instruction is the medical record quality control instruction, and the second instruction is the structural defect instruction. And acquiring data to be optimized from the first text data, taking the data to be optimized and the second instruction as input of the GPT model, taking the defect data as output of the GPT model, and training the GPT model to obtain the defect model. And generating a defect sample through the defect model, and judging the defect sample by taking the data to be optimized as a standard sample so as to optimize the defect judging model, wherein the defect judging model is used for responding to a first instruction to control the quality of the first text data. The method improves the efficiency and accuracy of medical record quality control.
Description
Technical Field
The invention relates to the technical field of medical assistance, in particular to a medical record quality control optimization method, a medical record quality control optimization device, electronic equipment and a storage medium.
Background
In medical information, medical records often need to accurately reflect symptoms, treatment progress and health conditions of patients, and quality requirements of medical staff on the medical records are also becoming more stringent. Inaccurate or incomplete medical records may cause misdiagnosis and mistreatment, which threatens the health and safety of the patient. Therefore, through regular medical record evaluation and quality control, the problems in the medical record can be effectively found, and an effective improvement strategy is timely formulated, so that the improvement of medical record writing standards and quality by medical staff is promoted, and the effect of improving medical service quality is further achieved.
At present, the existing medical record quality control methods are mainly divided into two types, namely a medical record quality control method based on rules and a medical record quality control method based on a large language model. The medical record quality control method based on the rules needs to evaluate the quality of medical records by formulating a series of clinical rules and standards, is easy to be accepted and implemented clinically, is easy to be expressed electronically, and is commonly used for primary construction of medical record quality evaluation. However, medical record quality control rules required by the medical record quality control method based on the rules need to be defined by professionals with professional clinical knowledge, and developers often lack professional medical background, and in addition, the rules are fixed and stiff, so that the medical record quality control rules are difficult to adapt to changes in clinical practice. The medical record quality control method based on the large language model mainly designs the medical record quality control rule into corresponding prompt, namely prompt dialog box type question-answer training data, based on the existing large language model. However, the medical record quality control method based on the large language model has higher dependency on the basic model, takes more time to optimize the template corresponding to the rule, and is less controllable. Although the fine tuning of the open-cell model can make the final model result more controllable, more manual data marking is required, more factors need to be considered in manual marking, and the writing format of the answer needs to be adjusted repeatedly in the process of marking the answer.
In summary, the existing medical record quality control method is more limited by complex and diverse model training data, and the efficiency and accuracy of medical record quality control are reduced to a certain extent.
Disclosure of Invention
Accordingly, in order to solve the above-mentioned problems, it is necessary to provide a medical record quality control optimization method, apparatus, electronic device and storage medium capable of improving the medical record quality control efficiency and accuracy.
The invention provides a medical record quality control optimization method, which comprises the following steps:
Acquiring medical record quality control rules and first text data, wherein the medical record quality control rules comprise medical record quality control ranges and medical record quality control definitions, and the first text data are text data to be controlled;
converting the medical record quality control rule into a first instruction and a second instruction respectively, wherein the first instruction is the medical record quality control instruction, and the second instruction is a construction defect instruction;
Obtaining data to be optimized from the first text data, taking the data to be optimized and the second instruction as input of a GPT model, taking defect data as output of the GPT model, and training the GPT model to obtain a defect model;
And generating a defect sample through the defect model, and judging the defect sample by taking the data to be optimized as a standard sample so as to optimize a defect judging model, wherein the defect judging model is used for responding to the first instruction to control the quality of the first text data.
In one embodiment, the obtaining the data to be optimized from the first text data, taking the data to be optimized and the second instruction as input of a GPT model, taking defect data as output, and training the GPT model to obtain a defect model includes:
combining the second instruction with the data to be optimized to obtain combined data, wherein the combined data has a defect construction rule of the second instruction;
and constructing input data of the GPT model based on the combined data, wherein the input data of the GPT model is jointly formed by the data to be optimized and the defect construction rule of the second instruction.
In one embodiment, the obtaining the data to be optimized from the first text data, taking the data to be optimized and the second instruction as input of a GPT model, taking defect data as output, and training the GPT model to obtain a defect model includes:
Sending the second instruction to the GPT model to call the GPT model to construct defect data corresponding to the data to be optimized according to the defect construction rule of the second instruction;
and filtering the defect data to obtain filtered data.
In one embodiment, the obtaining the data to be optimized from the first text data, taking the data to be optimized and the second instruction as input of a GPT model, taking defect data as output, and training the GPT model to obtain a defect model, and further includes:
Taking the combined data as model input data, taking the filtered data as model output data, and constructing a training data set;
and training the GPT model based on the training data set to obtain the defect model.
In one embodiment, the generating a defect sample by the defect model, and discriminating the defect sample by using the data to be optimized as a standard sample to optimize a defect discriminating model includes:
Acquiring a training data set and a test data set in the defect model, and generating the defect sample based on the training data set and the test data set;
And taking the data to be optimized as input data of the defect judging model, taking the defect sample as output data of the defect judging model, and carrying out optimization training on the defect judging model.
In one embodiment, the generating a defect sample by the defect model, and discriminating the defect sample by using the data to be optimized as a standard sample to optimize a defect discriminating model includes:
Sending the first instruction to the optimized defect judging model;
and calling the optimized defect discrimination model to discriminate the data to be optimized according to the quality control rule of the first instruction so as to output discrimination results.
In one embodiment, the method further comprises:
judging whether the data to be optimized is defect data or not based on the judging result; if yes, then
Outputting a medical record quality control result of which the data to be optimized is defect data; if not, then
And outputting medical record quality control results of which the data to be optimized are standard data meeting the medical record quality control rules.
The invention also provides a medical record quality control optimizing device, which comprises:
The data acquisition module is used for acquiring medical record quality control rules and first text data, wherein the medical record quality control rules comprise medical record quality control ranges and medical record quality control definitions, and the first text data are text data to be quality controlled;
the instruction conversion module is used for converting the medical record quality control rule into a first instruction and a second instruction respectively, wherein the first instruction is a medical record quality control instruction, and the second instruction is a construction defect instruction;
The model training module is used for acquiring data to be optimized from the first text data, taking the data to be optimized and the second instruction as input of a GPT model, taking defect data as output of the GPT model, and training the GPT model to obtain a defect model;
And the model optimization module is used for generating a defect sample through the defect model, judging the defect sample by taking the data to be optimized as a standard sample so as to optimize a defect judging model, and the defect judging model is used for responding to the first instruction to control the quality of the first text data.
The invention also provides an electronic device, which comprises a memory and a processor, wherein the memory stores a computer program, and the processor realizes the medical record quality control optimization method according to any one of the above when executing the computer program.
The invention also provides a computer storage medium storing a computer program which when executed by a processor realizes the medical record quality control optimizing method according to any one of the above.
The invention also provides a computer program product comprising a computer program which, when executed by a processor, implements a medical record quality control optimization method as described in any one of the above.
According to the medical record quality control optimizing method, the medical record quality control optimizing device, the electronic equipment and the storage medium, the medical record quality control rule comprising the medical record quality control range and the medical record quality control definition and the text data to be quality controlled are obtained. And then, respectively converting medical record quality control rules into medical record quality control instructions and construction defect instructions, acquiring data to be optimized from text data to be quality controlled, taking the data to be optimized and the construction defect instructions as input of a GPT model, taking defect data as output of the GPT model, and training the GPT model to obtain the defect model. And finally, generating a defect sample through the defect model, judging the defect sample by taking the data to be optimized as a standard sample, so as to optimize the defect judging model, and performing quality control on the text data to be controlled by the optimized defect judging model in response to a medical record quality control instruction. The method optimizes the medical record quality control effect by actively constructing defects, avoids the problems of complicated rules and manual marks and insufficient recall, improves the flexibility of defining and executing medical rules to a certain extent, and improves the efficiency and accuracy of medical record quality control.
Drawings
In order to more clearly illustrate the invention or the technical solutions of the prior art, the following description will briefly explain the drawings used in the embodiments or the description of the prior art, and it is obvious that the drawings in the following description are some embodiments of the invention, and other drawings can be obtained according to the drawings without inventive effort for a person skilled in the art.
FIG. 1 is a schematic flow chart of a medical record quality control optimization method provided by the invention;
Fig. 2 is a schematic diagram of an overall process of medical record quality control optimization according to the medical record quality control optimization method in the embodiment provided by the invention;
FIG. 3 is a second flowchart of the medical record quality control optimizing method according to the present invention;
FIG. 4 is a third schematic flow chart of the medical record quality control optimizing method according to the present invention;
FIG. 5 is a schematic flow chart of a medical record quality control optimization method according to the present invention;
FIG. 6 is a flowchart of a medical record quality control optimization method according to the present invention;
FIG. 7 is a flowchart illustrating a medical record quality control optimization method according to the present invention;
FIG. 8 is a schematic diagram of a medical record quality control optimization method according to the present invention;
FIG. 9 is a schematic structural diagram of a medical record quality control optimizing apparatus provided by the invention;
fig. 10 is an internal structural diagram of a computer device provided by the present invention.
Detailed Description
For the purpose of making the objects, technical solutions and advantages of the embodiments of the present invention more apparent, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention, and it is apparent that the described embodiments are some embodiments of the present invention, but not all embodiments of the present invention. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
The medical record quality control optimization method, the medical record quality control optimization device, the electronic equipment and the storage medium are described below with reference to fig. 1-10.
As shown in fig. 1, in one embodiment, a medical record quality control optimization method includes the following steps:
Step S110, obtaining medical record quality control rules and first text data, wherein the medical record quality control rules comprise medical record quality control ranges and medical record quality control definitions, and the first text data is text data to be quality controlled.
Specifically, the server acquires medical record quality control rules including medical record quality control ranges and medical record quality control definitions and text data to be quality controlled, namely first text data.
Wherein, the medical record quality control range and the medical record quality control definition are defined by human.
Referring to fig. 2, in a specific embodiment, the medical record quality control optimizing method provided by the present invention first needs to prepare a medical record quality control rule R, including determining a range of medical record quality control rules to be executed and definition thereof, and determining text fields related to the quality control rules.
Step S120, the medical record quality control rule is respectively converted into a first instruction and a second instruction, wherein the first instruction is the medical record quality control instruction, and the second instruction is the structural defect instruction.
Specifically, the server converts the medical record quality control rule obtained in step S110 into a medical record quality control instruction and a structural defect instruction, that is, a first instruction and a second instruction, respectively.
In particular embodiments, as shown in connection with fig. 2, after the medical record quality control rule R is determined, the medical record quality control rule R is converted into a corresponding instruction (promt) P1, and the medical record quality control rule R is converted into an instruction (promt) P2 that constructs a defect.
Step S130, obtaining data to be optimized from the first text data, taking the data to be optimized and the second instruction as input of the GPT model, taking the defect data as output of the GPT model, and training the GPT model to obtain the defect model.
Specifically, the server obtains data to be optimized from the first text data obtained in step S110, takes the data to be optimized and the defect constructing instruction as input of the GPT model, takes the defect data as output of the GPT model, and trains the GPT model to obtain a corresponding defect model.
In a specific embodiment, as shown in fig. 2, after the P1 and P2 instruction conversion is completed, the data D to be optimized is prepared based on the text field defined in the medical record quality control rule R, and then the model M of the active construction defect, that is, the defect model, is trained based on the GPT model. In the model training process, the instruction P2 and the data D need to be combined together first to construct a model input of the GPT. And secondly, a GPT model is called to actively construct defects for the data D to be optimized according to the defect rule instruction P2, and defect data D1 is obtained. Then, the data D1 is manually reviewed (review) to filter out high quality data, and finally training of the defect model M is completed.
And step S140, generating a defect sample through the defect model, and judging the defect sample by taking the data to be optimized as a standard sample so as to optimize the defect judging model, wherein the defect judging model is used for responding to a first instruction to control the quality of the first text data.
Specifically, the server generates abundant defect samples through the defect model obtained in the step S130, and judges the defect samples by taking the data to be optimized D as standard samples, so as to realize the optimization training of the defect judging model, so that the optimized defect judging model can more efficiently and accurately respond to the medical record quality control instruction to control the quality of the text data to be controlled.
In a specific embodiment, after the training of the defect model M is completed, a rich defect sample D 'is generated based on the defect model M, and then the training optimization is performed on the defect discrimination model based on the defect sample D' and the normal sample D (which are the data D to be optimized).
In the embodiment, the effect of medical record quality control is optimized by actively constructing defects, the problems of inefficiency and insufficient recall based on rules and manual marking data are overcome, the flexibility, the integrity, the accuracy and the use efficiency of medical rule definition and execution are further improved, and meanwhile, the requirements of medical service actual use and rule update are also met.
According to the medical record quality control optimization method, the medical record quality control rule comprising the medical record quality control range and the medical record quality control definition and the text data to be quality controlled are obtained. And then, respectively converting medical record quality control rules into medical record quality control instructions and construction defect instructions, acquiring data to be optimized from text data to be quality controlled, taking the data to be optimized and the construction defect instructions as input of a GPT model, taking defect data as output of the GPT model, and training the GPT model to obtain the defect model. And finally, generating a defect sample through the defect model, judging the defect sample by taking the data to be optimized as a standard sample, so as to optimize the defect judging model, and performing quality control on the text data to be controlled by the optimized defect judging model in response to a medical record quality control instruction. The method optimizes the medical record quality control effect by actively constructing defects, avoids the problems of complicated rules and manual marks and insufficient recall, improves the flexibility of defining and executing medical rules to a certain extent, and improves the efficiency and accuracy of medical record quality control.
As shown in fig. 3, in an embodiment, the medical record quality control optimizing method provided by the present invention obtains data to be optimized from first text data, takes the data to be optimized and second instructions as input of a GPT model, takes defect data as output of the GPT model, trains the GPT model to obtain a defect model, and includes the following steps:
Step S310, combining the second instruction and the data to be optimized to obtain combined data, wherein the combined data has a defect construction rule of the second instruction.
Specifically, before training the defect model, the server first combines the structural defect instruction with the data to be optimized to obtain combined data, where the combined data has a defect structural rule for constructing the defect instruction.
Step S320, based on the combined data, constructing input data of the GPT model, wherein the input data of the GPT model is composed of the data to be optimized and defect construction rules of the second instruction.
Specifically, the server constructs input data of the GPT model based on the combined data obtained in step S310, where the input data of the GPT model is composed of the data to be optimized and the defect construction rule for constructing the defect instruction.
As shown in fig. 4, in an embodiment, the medical record quality control optimization method provided by the present invention obtains data to be optimized from first text data, takes the data to be optimized and second instructions as input of a GPT model, takes defect data as output of the GPT model, trains the GPT model to obtain a defect model, and specifically includes the following steps:
step S132, a second instruction is sent to the GPT model to call the GPT model to construct defect data corresponding to the data to be optimized according to the defect construction rule of the second instruction.
Specifically, in the training process of the defect model, a construction defect instruction is firstly required to be sent to the GPT model so as to call the GPT model to construct defect data corresponding to the data to be optimized according to the defect construction rule of the construction defect instruction.
Step S134, filtering the defect data to obtain filtered data.
Specifically, the server performs filtering processing on the defect data obtained in step S132, to obtain filtered defect data.
For example, in a specific embodiment, after the P1 and P2 instruction conversion is completed, the data D to be optimized is prepared based on the text field defined in the medical record quality control rule R, and then the model M of the actively constructed defect, that is, the defect model, is trained based on the GPT model. In the model training process, the instruction P2 and the data D need to be combined together first to construct a model input of the GPT. And secondly, a GPT model is called to actively construct defects for the data D to be optimized according to the defect rule instruction P2, and defect data D1 is obtained. Then, the data D1 is manually reviewed (review) to filter out high quality data, and finally training of the defect model M is completed.
As shown in fig. 5, in an embodiment, the medical record quality control optimization method provided by the present invention obtains data to be optimized from first text data, takes the data to be optimized and second instructions as input of a GPT model, takes defect data as output of the GPT model, trains the GPT model to obtain a defect model, and specifically further includes the following steps:
And step S136, taking the combined data as model input data, taking the filtered data as model output data, and constructing a training data set.
Specifically, the server uses the previously obtained combined data as model input data, and uses the defect data obtained after filtering in step S132 as model output data, to construct a training data set.
Step S138, training the GPT model based on the training data set to obtain a defect model.
Specifically, the server trains the GPT model based on the training data set constructed in step S136, to obtain a final defect model.
As shown in fig. 6, in one embodiment, the medical record quality control optimizing method provided by the present invention generates a defect sample through a defect model, and uses data to be optimized as a standard sample to judge the defect sample so as to optimize the defect judging model, and specifically includes the following steps:
In step S142, a training data set and a test data set in the defect model are acquired, and a defect sample is generated based on the training data set and the test data set.
Specifically, in the process of optimizing the defect judging model, the server firstly needs to acquire a training data set and a test data set in the defect model, and generates a defect sample based on the training data set and the test data set.
And S144, taking the data to be optimized as input data of the defect judging model, taking a defect sample as output data of the defect judging model, and carrying out optimization training on the defect judging model.
Specifically, the server uses the data to be optimized as input data of the defect discrimination model, and simultaneously uses the defect sample obtained in step S142 as output data of the defect discrimination model to perform optimization training on the defect discrimination model.
For example, in a specific embodiment, after the defect model M is trained, a rich defect sample D 'is generated based on the defect model M, and then the defect discrimination model is trained and optimized based on the defect sample D' and the normal sample D (which are the data D to be optimized).
As shown in fig. 7, in an embodiment, the medical record quality control optimizing method provided by the present invention generates a defect sample through a defect model, and uses data to be optimized as a standard sample to judge the defect sample so as to optimize the defect judging model, and specifically further includes the following steps:
step S146, a first instruction is sent to the optimized defect discrimination model.
Specifically, the server sends a medical record quality control instruction, namely a first instruction, to the optimized defect discrimination model.
And S148, invoking the optimized defect discrimination model to discriminate the data to be optimized according to the quality control rule of the first instruction so as to output discrimination results.
Specifically, the server calls the optimized defect discrimination model to respond to the medical record quality control instruction and discriminate the data to be optimized according to the quality control rule in the medical record quality control instruction so as to output a corresponding discrimination result.
As shown in fig. 8, in one embodiment, the medical record quality control optimization method provided by the present invention further includes the following steps:
step S810, judging whether the data to be optimized is defect data or not based on the judging result.
Specifically, the server determines whether the data to be optimized is defect data based on the previously obtained discrimination result.
Step S820, outputting medical record quality control results with the data to be optimized as defect data.
Specifically, when the determination result in step S810 is that the data to be optimized is defect data, the server outputs a medical record quality control result that the data to be optimized is defect data.
Step S830, outputting medical record quality control results with the data to be optimized as standard data satisfying the medical record quality control rule.
Specifically, when the judgment result in step S810 is that the data to be optimized does not belong to the defect data, the server outputs a medical record quality control result that the data to be optimized is standard data satisfying the medical record quality control rule.
The medical record quality control optimizing device provided by the invention is described below, and the medical record quality control optimizing device described below and the medical record quality control optimizing method described above can be correspondingly referred to each other.
As shown in fig. 9, in one embodiment, a medical record quality control optimization device includes a data acquisition module 910, an instruction conversion module 920, a model training module 930, and a model optimization module 940.
The data obtaining module 910 is configured to obtain a medical record quality control rule and first text data, where the medical record quality control rule includes a medical record quality control range and a medical record quality control definition, and the first text data is text data to be quality controlled.
The instruction conversion module 920 is configured to convert the medical record quality control rule into a first instruction and a second instruction, where the first instruction is a medical record quality control instruction, and the second instruction is a structural defect instruction.
The model training module 930 is configured to obtain data to be optimized from the first text data, take the data to be optimized and the second instruction as input of the GPT model, take the defect data as output of the GPT model, and train the GPT model to obtain the defect model.
The model optimization module 940 is configured to generate a defect sample through the defect model, and determine the defect sample by using the data to be optimized as a standard sample, so as to optimize a defect determination model, where the defect determination model is configured to perform quality control on the first text data in response to the first instruction.
In this embodiment, the medical record quality control optimizing apparatus provided by the present invention further includes a data combining module, configured to:
And combining the second instruction and the data to be optimized to obtain combined data, wherein the combined data has the defect construction rule of the second instruction.
Based on the combined data, constructing input data of a GPT model, wherein the input data of the GPT model consists of data to be optimized and defect construction rules of a second instruction.
In this embodiment, the medical record quality control optimizing device provided by the invention, the model training module is specifically configured to:
And sending the second instruction to the GPT model to call the GPT model to construct defect data corresponding to the data to be optimized according to the defect construction rule of the second instruction.
And filtering the defect data to obtain filtered data.
In this embodiment, the medical record quality control optimizing device provided by the invention, the model training module is specifically further configured to:
And constructing a training data set by taking the combined data as model input data and the filtered data as model output data.
And training the GPT model based on the training data set to obtain a defect model.
In this embodiment, the medical record quality control optimizing device provided by the invention, the model optimizing module is specifically configured to:
A training data set and a test data set in the defect model are acquired, and a defect sample is generated based on the training data set and the test data set.
And taking the data to be optimized as input data of the defect discrimination model, taking a defect sample as output data of the defect discrimination model, and carrying out optimization training on the defect discrimination model.
In this embodiment, the medical record quality control optimizing apparatus provided by the present invention further includes a data discriminating module, configured to:
and sending the first instruction to the optimized defect judging model.
And calling the optimized defect discrimination model to discriminate the data to be optimized according to the quality control rule of the first instruction so as to output discrimination results.
In this embodiment, the medical record quality control optimizing apparatus provided by the present invention further includes a result judging module, configured to:
And judging whether the data to be optimized is defect data or not based on the judging result. If yes, then
And outputting a medical record quality control result of which the data to be optimized is defect data. If not, then
And outputting medical record quality control results of which the data to be optimized is standard data meeting medical record quality control rules.
Fig. 10 illustrates a physical structure diagram of an electronic device, which may be an intelligent terminal, and an internal structure diagram thereof may be as shown in fig. 10. The electronic device includes a processor, a memory, and a network interface connected by a system bus. Wherein the processor of the electronic device is configured to provide computing and control capabilities. The memory of the electronic device includes a nonvolatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage media. The network interface of the electronic device is used for communicating with an external terminal through a network connection. The computer program, when executed by a processor, implements a medical record quality control optimization method, the method comprising:
acquiring medical record quality control rules and first text data, wherein the medical record quality control rules comprise medical record quality control ranges and medical record quality control definitions, and the first text data are text data to be quality controlled;
Converting the medical record quality control rule into a first instruction and a second instruction respectively, wherein the first instruction is the medical record quality control instruction, and the second instruction is the structural defect instruction;
Obtaining data to be optimized from the first text data, taking the data to be optimized and the second instruction as input of a GPT model, taking defect data as output of the GPT model, and training the GPT model to obtain a defect model;
and generating a defect sample through the defect model, and judging the defect sample by taking the data to be optimized as a standard sample so as to optimize the defect judging model, wherein the defect judging model is used for responding to a first instruction to control the quality of the first text data.
It will be appreciated by those skilled in the art that the structure shown in fig. 10 is merely a block diagram of a portion of the structure associated with the present inventive arrangements and is not limiting of the electronic device to which the present inventive arrangements are applied, and that a particular electronic device may include more or fewer components than shown, or may combine certain components, or have a different arrangement of components.
In another aspect, the present invention further provides a computer storage medium storing a computer program, where the computer program when executed by a processor implements a medical record quality control optimization method, the method includes:
acquiring medical record quality control rules and first text data, wherein the medical record quality control rules comprise medical record quality control ranges and medical record quality control definitions, and the first text data are text data to be quality controlled;
Converting the medical record quality control rule into a first instruction and a second instruction respectively, wherein the first instruction is the medical record quality control instruction, and the second instruction is the structural defect instruction;
Obtaining data to be optimized from the first text data, taking the data to be optimized and the second instruction as input of a GPT model, taking defect data as output of the GPT model, and training the GPT model to obtain a defect model;
and generating a defect sample through the defect model, and judging the defect sample by taking the data to be optimized as a standard sample so as to optimize the defect judging model, wherein the defect judging model is used for responding to a first instruction to control the quality of the first text data.
In yet another aspect, a computer program product or computer program is provided, the computer program product or computer program comprising computer instructions stored in a computer readable storage medium. A processor of an electronic device reads the computer instructions from a computer readable storage medium, and when the processor executes the computer instructions, the processor implements a medical record quality control optimization method, the method comprising:
acquiring medical record quality control rules and first text data, wherein the medical record quality control rules comprise medical record quality control ranges and medical record quality control definitions, and the first text data are text data to be quality controlled;
Converting the medical record quality control rule into a first instruction and a second instruction respectively, wherein the first instruction is the medical record quality control instruction, and the second instruction is the structural defect instruction;
Obtaining data to be optimized from the first text data, taking the data to be optimized and the second instruction as input of a GPT model, taking defect data as output of the GPT model, and training the GPT model to obtain a defect model;
and generating a defect sample through the defect model, and judging the defect sample by taking the data to be optimized as a standard sample so as to optimize the defect judging model, wherein the defect judging model is used for responding to a first instruction to control the quality of the first text data.
Those skilled in the art will appreciate that implementing all or part of the above described methods may be accomplished by way of a computer program stored on a non-transitory computer readable storage medium, which when executed, may comprise the steps of the embodiments of the methods described above. Any reference to memory, storage, database, or other medium used in embodiments provided herein may include non-volatile and/or volatile memory. The nonvolatile memory can include Read Only Memory (ROM), programmable ROM (PROM), electrically Programmable ROM (EPROM), electrically Erasable Programmable ROM (EEPROM), or flash memory. Volatile memory can include Random Access Memory (RAM) or external cache memory.
By way of illustration and not limitation, RAM is available in a variety of forms such as Static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double Data Rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (SYNCHLINK) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), among others.
The technical features of the above embodiments may be arbitrarily combined, and all possible combinations of the technical features in the above embodiments are not described for brevity of description, however, as long as there is no contradiction between the combinations of the technical features, they should be considered as the scope of the description.
The foregoing examples illustrate only a few embodiments of the invention and are described in detail herein without thereby limiting the scope of the invention. It should be noted that it will be apparent to those skilled in the art that several variations and modifications can be made without departing from the spirit of the invention, which are all within the scope of the invention. Accordingly, the scope of protection of the present invention is to be determined by the appended claims.
Claims (10)
1. A medical record quality control optimization method, which is characterized by comprising the following steps:
Acquiring medical record quality control rules and first text data, wherein the medical record quality control rules comprise medical record quality control ranges and medical record quality control definitions, and the first text data are text data to be controlled;
converting the medical record quality control rule into a first instruction and a second instruction respectively, wherein the first instruction is the medical record quality control instruction, and the second instruction is a construction defect instruction;
Obtaining data to be optimized from the first text data, taking the data to be optimized and the second instruction as input of a GPT model, taking defect data as output of the GPT model, and training the GPT model to obtain a defect model;
And generating a defect sample through the defect model, and judging the defect sample by taking the data to be optimized as a standard sample so as to optimize a defect judging model, wherein the defect judging model is used for responding to the first instruction to control the quality of the first text data.
2. The medical record quality control optimization method according to claim 1, wherein the obtaining the data to be optimized from the first text data, taking the data to be optimized and the second instruction as input of a GPT model, taking defect data as output, training the GPT model to obtain a defect model, and the steps of:
combining the second instruction with the data to be optimized to obtain combined data, wherein the combined data has a defect construction rule of the second instruction;
and constructing input data of the GPT model based on the combined data, wherein the input data of the GPT model is jointly formed by the data to be optimized and the defect construction rule of the second instruction.
3. The medical record quality control optimization method according to claim 2, wherein the obtaining the data to be optimized from the first text data, taking the data to be optimized and the second instruction as input of a GPT model, taking defect data as output, training the GPT model to obtain a defect model, includes:
Sending the second instruction to the GPT model to call the GPT model to construct defect data corresponding to the data to be optimized according to the defect construction rule of the second instruction;
and filtering the defect data to obtain filtered data.
4. The medical record quality control optimizing method according to claim 3, wherein the obtaining the data to be optimized from the first text data, taking the data to be optimized and the second instruction as input of a GPT model, taking defect data as output, training the GPT model to obtain a defect model, further comprises:
Taking the combined data as model input data, taking the filtered data as model output data, and constructing a training data set;
and training the GPT model based on the training data set to obtain the defect model.
5. The medical record quality control optimizing method according to claim 1, wherein the generating a defect sample by the defect model and discriminating the defect sample with the data to be optimized as a standard sample to optimize a defect discriminating model includes:
Acquiring a training data set and a test data set in the defect model, and generating the defect sample based on the training data set and the test data set;
And taking the data to be optimized as input data of the defect judging model, taking the defect sample as output data of the defect judging model, and carrying out optimization training on the defect judging model.
6. The medical record quality control optimization method according to claim 1, wherein the generating a defect sample by the defect model and discriminating the defect sample by using the data to be optimized as a standard sample to optimize a defect discriminating model comprises:
Sending the first instruction to the optimized defect judging model;
and calling the optimized defect discrimination model to discriminate the data to be optimized according to the quality control rule of the first instruction so as to output discrimination results.
7. The medical record quality control optimization method according to claim 6, further comprising:
judging whether the data to be optimized is defect data or not based on the judging result; if yes, then
Outputting a medical record quality control result of which the data to be optimized is defect data; if not, then
And outputting medical record quality control results of which the data to be optimized are standard data meeting the medical record quality control rules.
8. A medical record quality control optimizing apparatus, characterized in that the apparatus comprises:
The data acquisition module is used for acquiring medical record quality control rules and first text data, wherein the medical record quality control rules comprise medical record quality control ranges and medical record quality control definitions, and the first text data are text data to be quality controlled;
the instruction conversion module is used for converting the medical record quality control rule into a first instruction and a second instruction respectively, wherein the first instruction is a medical record quality control instruction, and the second instruction is a construction defect instruction;
The model training module is used for acquiring data to be optimized from the first text data, taking the data to be optimized and the second instruction as input of a GPT model, taking defect data as output of the GPT model, and training the GPT model to obtain a defect model;
And the model optimization module is used for generating a defect sample through the defect model, judging the defect sample by taking the data to be optimized as a standard sample so as to optimize a defect judging model, and the defect judging model is used for responding to the first instruction to control the quality of the first text data.
9. An electronic device comprising a memory and a processor, the memory storing a computer program, characterized in that the processor implements the steps of the method of any one of claims 1 to 7 when the computer program is executed.
10. A computer storage medium storing a computer program, characterized in that the computer program when executed by a processor implements the steps of the method of any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410406561.5A CN117995346B (en) | 2024-04-07 | 2024-04-07 | Medical record quality control optimization method and device, electronic equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410406561.5A CN117995346B (en) | 2024-04-07 | 2024-04-07 | Medical record quality control optimization method and device, electronic equipment and storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117995346A true CN117995346A (en) | 2024-05-07 |
| CN117995346B CN117995346B (en) | 2024-06-25 |
Family
ID=90891063
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410406561.5A Active CN117995346B (en) | 2024-04-07 | 2024-04-07 | Medical record quality control optimization method and device, electronic equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117995346B (en) |
Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110853718A (en) * | 2019-10-16 | 2020-02-28 | 云知声智能科技股份有限公司 | Intelligent medical record quality control system and method |
| WO2021114620A1 (en) * | 2020-06-16 | 2021-06-17 | 平安科技（深圳）有限公司 | Medical-record quality control method, apparatus, computer device, and storage medium |
| WO2021151302A1 (en) * | 2020-07-16 | 2021-08-05 | 平安科技（深圳）有限公司 | Drug quality-control analysis method, apparatus, device, and medium based on machine learning |
| CN113435200A (en) * | 2021-06-22 | 2021-09-24 | 上海交通大学医学院附属仁济医院 | Entity recognition model training and electronic medical record processing method, system and equipment |
| CN113590396A (en) * | 2021-07-23 | 2021-11-02 | 南方电网深圳数字电网研究院有限公司 | Method and system for diagnosing defect of primary device, electronic device and storage medium |
| CN114240160A (en) * | 2021-12-17 | 2022-03-25 | 挂号网（杭州）科技有限公司 | Quality control method and device for electronic medical record, electronic equipment and storage medium |
| WO2022174491A1 (en) * | 2021-02-19 | 2022-08-25 | 平安科技（深圳）有限公司 | Artificial intelligence-based method and apparatus for medical record quality control, computer device, and storage medium |
| CN116682524A (en) * | 2023-07-17 | 2023-09-01 | 北京惠每云科技有限公司 | Medical record quality control method and device, electronic equipment and storage medium |
| CN117594173A (en) * | 2023-11-21 | 2024-02-23 | 广东阳普智慧医疗信息科技有限公司 | Training method and device for traditional Chinese medical record model |
| CN117809826A (en) * | 2024-02-28 | 2024-04-02 | 山东佰泰丰信息科技有限公司 | Medical record quality control method based on large model |
-
2024
- 2024-04-07 CN CN202410406561.5A patent/CN117995346B/en active Active
Patent Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110853718A (en) * | 2019-10-16 | 2020-02-28 | 云知声智能科技股份有限公司 | Intelligent medical record quality control system and method |
| WO2021114620A1 (en) * | 2020-06-16 | 2021-06-17 | 平安科技（深圳）有限公司 | Medical-record quality control method, apparatus, computer device, and storage medium |
| WO2021151302A1 (en) * | 2020-07-16 | 2021-08-05 | 平安科技（深圳）有限公司 | Drug quality-control analysis method, apparatus, device, and medium based on machine learning |
| WO2022174491A1 (en) * | 2021-02-19 | 2022-08-25 | 平安科技（深圳）有限公司 | Artificial intelligence-based method and apparatus for medical record quality control, computer device, and storage medium |
| CN113435200A (en) * | 2021-06-22 | 2021-09-24 | 上海交通大学医学院附属仁济医院 | Entity recognition model training and electronic medical record processing method, system and equipment |
| CN113590396A (en) * | 2021-07-23 | 2021-11-02 | 南方电网深圳数字电网研究院有限公司 | Method and system for diagnosing defect of primary device, electronic device and storage medium |
| CN114240160A (en) * | 2021-12-17 | 2022-03-25 | 挂号网（杭州）科技有限公司 | Quality control method and device for electronic medical record, electronic equipment and storage medium |
| CN116682524A (en) * | 2023-07-17 | 2023-09-01 | 北京惠每云科技有限公司 | Medical record quality control method and device, electronic equipment and storage medium |
| CN117594173A (en) * | 2023-11-21 | 2024-02-23 | 广东阳普智慧医疗信息科技有限公司 | Training method and device for traditional Chinese medical record model |
| CN117809826A (en) * | 2024-02-28 | 2024-04-02 | 山东佰泰丰信息科技有限公司 | Medical record quality control method based on large model |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117995346B (en) | 2024-06-25 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109659013B (en) | Disease diagnosis and path optimization method, device, equipment and storage medium | |
| CN109887596A (en) | Chronic obstructive disease of lung diagnostic method, device and the computer equipment of knowledge based map | |
| CN117995373B (en) | Hospital ward record generation method and device, electronic equipment and storage medium | |
| CN109840280B (en) | Text classification method and device and computer readable storage medium | |
| CN116629227B (en) | Method and equipment for converting text into SQL (structured query language) sentence | |
| CN114547274B (en) | Multi-turn question and answer method, device and equipment | |
| CN117995340B (en) | Intelligent recruitment method and device for clinical trial based on large model | |
| CN116610819A (en) | Medical knowledge graph generation method and device, electronic equipment and storage medium | |
| CN113782192A (en) | Grouping model construction method based on causal inference and medical data processing method | |
| CN117995346B (en) | Medical record quality control optimization method and device, electronic equipment and storage medium | |
| CN110134775B (en) | Question and answer data generation method and device and storage medium | |
| CN115691742B (en) | Electronic medical record quality control method, system, storage medium and equipment | |
| CN117194632A (en) | Method, device, equipment and medium for extracting structured knowledge from document | |
| CN117743522A (en) | Medical reply generation method, device, equipment and storage medium | |
| CN116862017A (en) | Data processing method, device, equipment and medium | |
| CN114743693A (en) | Doctor-patient dialogue based center quality control method and center quality control device | |
| KR102306953B1 (en) | Method of medical support based on voice recognition | |
| CN114462376A (en) | RPA and AI-based court trial record generation method, device, equipment and medium | |
| CN114341867A (en) | Translation method, translation device, client, server and storage medium | |
| Van der Elst | Assumptions of the Normal Error Regression Model | |
| CN117219247B (en) | Intelligent management system for patient treatment | |
| CN117409432A (en) | Image information extraction method, model training method, device, equipment and medium | |
| CN118039138A (en) | Controllable path pre-consultation method and device based on large model | |
| CN118471550A (en) | Intelligent medical assistant system based on large language model, interaction method thereof and product | |
| Lin et al. | Chain of Condition: Construct, Verify and Solve Conditions for Conditional Question Answering |
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 |