Emergency diagnosis and treatment information page generation method, RPA process design method and medium

Abstract

The invention relates to an outpatient and emergency diagnosis and treatment information page generation method, an RPA process design method and a medium, and relates to the technical field of medical information extraction. In the scheme, based on a first RPA process and a third RPA process, structural data in an HIS system and an electronic medical record system are extracted and stored in an information page database; extracting unstructured data in the HIS system and the electronic medical record system based on the second RPA process and the fourth RPA process, extracting a first entity object and a second entity object based on the first entity object neural network identification model and the second entity object neural network identification model, and storing the first entity object and the second entity object in an information page database; and performing correspondence, matching and editing on the data in the information page database and the data items in the structured information page to obtain the information page corresponding to the patient. Compared with the prior art, the method can be suitable for various versions of systems, is better in compatibility, is low in development cost, and is higher in market value and economic value.

Description

Technical Field

[0001] The present invention relates to the technical field of medical information extraction, and in particular to a method for generating an outpatient and emergency treatment information page, an RPA process design method and a medium. Background Art

[0002] The outpatient (emergency) diagnosis and treatment information page (hereinafter referred to as the "information page", a total of 72 fields) is a summary of the patient's current visit process, which is formed by the hospital based on the outpatient (emergency) medical records and various information generated by the patient during the outpatient (emergency) visit of the hospital, including the patient's basic information, visit process information, diagnosis and treatment information, and cost information. In order to further strengthen medical quality management, improve the refinement and informatization level of outpatient (emergency) medical quality management, and promote the improvement of outpatient (emergency) medical quality, the automatic generation of information pages has become a problem that needs to be solved urgently. If the current hospitals can realize the automatic generation of information pages in a timely and accurate manner, it can greatly improve the refinement and informatization level of outpatient (emergency) medical management.

[0003] However, in the current information extraction of information pages, it is necessary to open up a data interface, which faces high data interface fees and development costs. Summary of the invention

[0004] The technical problem to be solved by this application is to provide a method for generating outpatient and emergency treatment information pages, an RPA process design method and a medium, which are applicable to multiple version systems, have better compatibility and lower cost.

[0005] In a first aspect, an embodiment provides a method for generating an outpatient and emergency department treatment information page, comprising: Based on the first RPA process, the structured data in the HIS system is extracted and stored in the information page database; Based on the second RPA process, unstructured data in the HIS system is extracted, and based on the extracted unstructured data in the HIS system, the trained first entity object neural network recognition model is used to extract the first entity object in the unstructured data, and the extracted first entity object is stored in the information page database; Based on the third RPA process, the structured data in the electronic medical record system is extracted and stored in the information page database; Based on the fourth RPA process, unstructured data in the electronic medical record system is extracted, and based on the extracted unstructured data in the electronic medical record system, the trained second entity object neural network recognition model is used to extract the second entity object in the unstructured data, and the extracted second entity object is stored in the information page database; The data in the information page database are corresponded, matched and edited with the data items in the structured information page to obtain the information page corresponding to each patient.

[0006] In one embodiment, a design method of the first RPA process and / or the third RPA process includes: Determine the structured data that needs to be extracted as the target structured data; Simulate the operation of manually logging into the HIS system and / or electronic medical record system; Navigate to the page containing the target structured data through manual operation; Based on data crawling or reading functions, target structured data is extracted according to the preset location and format.

[0007] In one embodiment, the design method of the first RPA process and / or the third RPA process further includes: Perform data conversion and cleaning on the extracted data.

[0008] In one embodiment, the design method of the second RPA process and / or the fourth RPA process includes: Determine the specific business scenarios of the target unstructured data that needs to be extracted in the HIS system and / or electronic medical record system, and determine the type, source, and specific fields or information of the target unstructured data to be extracted; Simulate the operation of manually logging into the HIS system and / or electronic medical record system; Navigate to the page or document containing the target unstructured data through manual operation; According to the characteristics and extraction requirements of the target unstructured data, set corresponding data extraction rules to extract the target unstructured data.

[0009] In one embodiment, the training method of the first entity object neural network recognition model includes: Collecting a plurality of unstructured medical text data, wherein the medical text data includes a first entity object to be extracted; Annotating the medical text data with a first entity object that needs to be extracted; Convert the annotated medical text data into a format suitable for the Hua Tuo model input to obtain annotated data; Dividing the labeled data into a training set, a validation set, and a test set; Based on the preset training parameters, the training set is used as the input of the Hua Tuo model for training. The performance of the Hua Tuo model is evaluated and the hyperparameters are adjusted based on the validation set during the training process. The generalization ability of the Hua Tuo model is evaluated based on the test set, thereby obtaining a trained and improved Hua Tuo model. The trained improved Hua Tuo model is used as the first entity object neural network recognition model.

[0010] In one embodiment, the training method of the second entity object neural network recognition model includes: Collecting a plurality of unstructured medical text data, wherein the medical text data includes a second entity object to be extracted; Annotating the medical text data with a second entity object that needs to be extracted; Convert the annotated medical text data into a format suitable for the Hua Tuo model input to obtain annotated data; Dividing the labeled data into a training set, a validation set, and a test set; Based on the preset training parameters, the training set is used as the input of the Hua Tuo model for training. The performance of the Hua Tuo model is evaluated and the hyperparameters are adjusted based on the validation set during the training process. The generalization ability of the Hua Tuo model is evaluated based on the test set, thereby obtaining a trained and improved Hua Tuo model. The trained improved Hua Tuo model is used as the second entity object neural network recognition model.

[0011] In one embodiment, the corresponding, matching and editing of the data in the information page database with the data items in the structured information page includes: corresponding to the corresponding data items, and matching the corresponding data item options to edit and select the matched options.

[0012] In one embodiment, the method further includes: saving the generated information page by naming it as patient name + clinic time.

[0013] In a second aspect, an embodiment provides an RPA process design method for obtaining an RPA process in the information page generation method described in any one of the above embodiments, including obtaining a first RPA process, a second RPA process, a third RPA process, and a fourth RPA process; the design method includes: Obtaining the content that needs to be edited on the information page and classifying it, wherein the classification includes the content that needs to be extracted from the HIS system and the content that needs to be extracted from the electronic medical record system; Obtaining a classification of content that needs to be extracted from the HIS system, and further classifying the content into first structured data and first unstructured data; Based on the first structured data, a first RPA process is designed, and based on the first unstructured data, a second RPA process is designed; Obtaining a classification of content that needs to be extracted from the electronic medical record system, and further classifying the content into second structured data and second unstructured data; Based on the second structured data, a third RPA process is designed, and based on the fourth unstructured data, a fourth RPA process is designed.

[0014] In a third aspect, an embodiment provides a computer-readable storage medium, in which a program is stored, and the program can be loaded by a processor to execute the information page generation method and / or RPA process design method described in any one of the above embodiments.

[0015] The beneficial effects of the present invention are: Based on the designed RPA process, the structured data and unstructured data required for the information page are extracted. Since only the authorization of login and data reading permissions is required, there is no need to open the manufacturer's data interface, nor is there a need to customize the integration platform, and there is no need to consider compatibility issues. In this process, the data interface fee is saved, and the link of opening the data interface fee is saved. On the one hand, the solution of the embodiment of the present application can be applied to multiple versions of systems, with better compatibility and low development cost. On the other hand, once the HIS system of the mainstream HIS manufacturer and the electronic medical record system of the electronic medical record manufacturer on the market are accurately adapted once, then in the future, if other hospitals use the adapted HIS system and electronic medical record system, they do not need to be adapted again. This deploys a software robot that automatically generates information pages in this hospital, which can greatly save deployment time. Therefore, it has high market value and economic value. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 This is a flowchart of an RPA process design method according to an embodiment of the present application; Figure 2 This is a flowchart of a first RPA process design method according to an embodiment of the present application; Figure 3 This is a flowchart of a second RPA process design method according to an embodiment of the present application; Figure 4 This is a flowchart of a method for generating an outpatient and emergency treatment information page according to an embodiment of the present application; Figure 5 It is a flowchart of a training method for a first entity object neural network recognition model according to an embodiment of the present application. DETAILED DESCRIPTION

[0017] The present invention is further described in detail below by specific embodiments in conjunction with the accompanying drawings. Wherein similar elements in different embodiments adopt associated similar element numbers. In the following embodiments, many detailed descriptions are for making the present application better understood. However, those skilled in the art can easily recognize that some features can be omitted in different situations, or can be replaced by other elements, materials, methods. In some cases, some operations related to the present application are not shown or described in the specification, this is to avoid the core part of the present application being overwhelmed by too much description, and for those skilled in the art, it is not necessary to describe these related operations in detail, and they can fully understand the related operations according to the description in the specification and the general technical knowledge in the art.

[0018] In addition, the features, operations or characteristics described in the specification can be combined in any appropriate manner to form various implementations. At the same time, the steps or actions in the method description can also be interchanged or adjusted in a manner that is obvious to those skilled in the art. Therefore, the various sequences in the specification and the drawings are only for the purpose of clearly describing a certain embodiment and are not meant to be a required sequence, unless otherwise specified that a certain sequence must be followed.

[0019] The serial numbers assigned to the components in this article, such as "first", "second", etc., are only used to distinguish the objects described and do not have any order or technical meaning.

[0020] To facilitate the description of the inventive concept of the present application, the information page generation technology is briefly described below.

[0021] The information on the information page mainly comes from two systems in the hospital: the HIS (hospital information system) system and the electronic medical record system. With current technology, for hospitals that have deployed an integrated platform, data can be extracted through the integrated platform, and then the information page template can be customized to inject data to generate an information page. However, there are currently few hospitals that have deployed an integrated platform, and the integrated platform requires customized development, has poor compatibility, and has high development costs and data interface connection costs. For hospitals that have not deployed an integrated platform, it is necessary to extract data from the hospital's HIS system and electronic medical record system respectively. However, this requires opening a data interface with the system manufacturer, which requires paying a large amount of data interface fees and development costs.

[0022] In view of this, an embodiment of the present application provides an outpatient and emergency treatment information page generation method, an RPA (robotic process automation) process design method and a medium. In the scheme, based on the designed first RPA process and the third RPA process, the structured data in the HIS system and the electronic medical record system required for the information page are extracted respectively, and based on the designed second RPA process and the fourth RPA process, the unstructured data in the HIS system and the electronic medical record system required for the information page are extracted respectively. Then, based on the unstructured data, the corresponding neural network model is used to extract the corresponding entity object, so that all the data required for the information page can be obtained and the information page is generated. Since the process is based on the RPA process, there is no need to open up the manufacturer's data interface, so that the data interface fee can be saved, and the link of opening up the data interface fee can be saved. There is no need to customize the integration platform, and there is no need to consider the compatibility issues between the various systems in the integration platform. Therefore, the embodiment of the present application can be applicable to multiple versions of systems, with better compatibility and low development cost, and has high market and economic value.

[0023] In order to more clearly illustrate the embodiment of the present application, the RPA process design method is first described below.

[0024] In one embodiment of the present application, an RPA process design method is provided, which can be used to obtain the RPA process in the outpatient and emergency treatment information page generation method. Please refer to Figure 1 , the design methods include: Step S10, obtaining the content to be edited on the information page and classifying it, including the content to be extracted from the HIS system and the content to be extracted from the electronic medical record system.

[0025] In the current information page, there are 72 fields that need to be obtained, including structured data and unstructured data (such as allergy history). Some of the 72 fields need to be obtained from the HIS system, and some need to be obtained from the electronic medical record. Therefore, the content that needs to be edited is divided into two categories: content that needs to be extracted from the HIS system and content that needs to be extracted from the electronic medical record system.

[0026] In one embodiment, the information page may be demarcated by marking boxes for selection and classification.

[0027] Step S20, obtaining a classification of content to be extracted from the HIS system, and further classifying the content into first structured data and first unstructured data.

[0028] Step S30: designing a first RPA process based on the first structured data, and designing a second RPA process based on the first unstructured data.

[0029] In one embodiment, please refer to Figure 2 , the design method of the first RPA process can include: Step S3011: determine the structured data to be extracted as target structured data.

[0030] The target structured data may be the patient's basic information, diagnosis results, doctor's advice information, examination and test reports, etc. It is necessary to determine which specific structured data needs to be extracted. In this regard, in one embodiment, the structured data to be extracted by the first RPA process may be extracted based on the structured data to be extracted from the HIS system as defined in the information page.

[0031] Step S3012, simulating the operation of manually logging into the HIS system.

[0032] Simulate the operation of manually logging into the HIS system and enter login information such as user name and password.

[0033] Step S3013: Navigate to a page containing target structured data through manual operation.

[0034] Navigate to the page containing the target structured data by clicking a menu, button, or entering a query.

[0035] Step S3014, based on the data crawling or reading function, extract the target structured data according to the preset position and format.

[0036] Use data crawling or reading functions (for example, using text extraction, table reading, etc. to obtain data) to extract the required target structured data according to a predetermined location and format.

[0037] In one embodiment, the extracted target structured data is stored in a pre-set information page database. In one embodiment, the information page database is an information page data table.

[0038] In one embodiment, before storing the extracted target structured data in a pre-set information page database, the extracted data is further converted and cleaned. In one embodiment, the data conversion and cleaning process includes unifying the data format, removing spaces, invalid characters, etc., to ensure the quality and consistency of the data.

[0039] In one embodiment, please refer to Figure 3 , the design methods of the second RPA process can include: Step S3021, determine the specific business scenario of the target unstructured data to be extracted in the HIS system, and determine the type, source, and specific fields or information of the target unstructured data to be extracted.

[0040] Determine the specific business scenarios of the target unstructured data that needs to be extracted in the HIS system, such as the extraction of patient diagnosis reports, analysis of medical order records, etc., so as to clarify the business process.

[0041] Determine the type, source, and specific fields or information of the target unstructured data to be extracted, such as text descriptions, images, etc., as well as the storage location and access method of this data in the system to define the data scope.

[0042] Step S3022, simulating the operation of manually logging into the HIS system.

[0043] Simulate the operation of manually logging into the HIS system, configure the connection with the HIS system through the RPA tool, set access rights and authentication information, and ensure that the RPA robot can legally log in and access the target system.

[0044] Step S3023: Navigate to the page or document containing the target unstructured data through manual operation.

[0045] Use the interface operation recording or element location function of the RPA tool to determine the specific steps and methods for finding and locating the target unstructured data in the system. For example, navigate to the page or document containing the target unstructured data by specifying a menu path, entering keywords in the search box, clicking a button, and so on.

[0046] Step S3024, according to the characteristics of the target unstructured data and the extraction requirements, set corresponding data extraction rules to extract the target unstructured data.

[0047] For example, extracting a document, a page, or one or more paragraphs of content that meet the extraction rules.

[0048] Step S40, obtaining a classification of the content to be extracted from the electronic medical record system, and further classifying the content into second structured data and second unstructured data.

[0049] Step S50: design a third RPA process based on the second structured data, and design a fourth RPA process based on the fourth unstructured data.

[0050] In one embodiment, the third RPA process may be designed with reference to steps S3011 to S3014. In one embodiment, the fourth RPA process may be designed with reference to steps S3021 to S3024.

[0051] Those skilled in the art can understand that when designing the third RPA process with reference to steps S3011 to S3014, the login object is the electronic medical record system; when designing the fourth RPA process with reference to steps S3021 to S3024, the login object is the electronic medical record system.

[0052] Based on the above embodiments, it can be known that the structured data and unstructured data required for the information page are extracted based on the designed RPA process. Since only authorization such as login and data reading permissions is required, there is no need to open up the manufacturer's data interface, nor is there a need to customize the integration platform, and there is no need to consider compatibility issues. In this process, the data interface fee is saved, and the link of opening up the data interface fee is saved, so that the solution of the embodiment of the present application can be applied to multiple versions of systems, with better compatibility and low development cost, and has high market and economic value.

[0053] In one embodiment of the present application, a method for generating an outpatient and emergency treatment information page is provided, which can be implemented based on the RPA process design method provided in the above embodiment. Please refer to Figure 4 , the information page generation method includes: Step S100: extract structured data from the HIS system based on the first RPA process and store it in the information page database.

[0054] Step S200: Based on the second RPA process, extract the unstructured data in the HIS system, and based on the extracted unstructured data in the HIS system, use the trained first entity object neural network recognition model to extract the first entity object in the unstructured data, and store the extracted first entity object in the information page database.

[0055] Based on the RPA process, the first entity object can be extracted from a certain document, a certain page, or one or more paragraphs of content. However, in combination with the RPA process, how to obtain the first entity object of the unstructured data required for the information page from the extracted document, a certain page, or one or more paragraphs of content is a technical difficulty.

[0056] In view of this, in one embodiment of the present application, based on the Hua Tuo model suitable for the medical field, the Hua Tuo model is trained and fine-tuned to obtain a neural network model for extracting entity objects suitable for embedding in the RPA process. Figure 5 , the training method of the first entity object neural network recognition model may include: Step S2001: Collect a plurality of unstructured medical text data, wherein the medical text data includes a first entity object (keyword object) to be extracted.

[0057] In one embodiment, the method further includes cleaning and preprocessing the collected multiple structured medical text data to remove noise, duplicate data and irrelevant information.

[0058] Step S2002: annotate the first entity object to be extracted from the medical text data.

[0059] Through labeling, the first entity object that needs to be extracted by the model can be clearly pointed out. Labeling requires professional medical knowledge and a rigorous attitude to ensure the accuracy and consistency of labeling. For example, labeling entity information such as symptoms, diseases, and drugs in medical records.

[0060] Step S2003, converting the annotated medical text data into a format suitable for input into the Hua Tuo model to obtain annotated data.

[0061] Those skilled in the art can understand that text data needs to be converted into a vector representation that can be understood by the model through word segmentation, word vector conversion, etc.

[0062] Step S2004, dividing the labeled data into a training set, a validation set and a test set.

[0063] The training set is used to train the model, the validation set is used to evaluate the performance of the model and adjust the hyperparameters during the training process, and the test set is used to finally evaluate the generalization ability of the model.

[0064] Step S2005, based on the preset training parameters, the training set is used as the input of the Hua Tuo model for training, the performance of the Hua Tuo model is evaluated and the hyperparameters are adjusted based on the validation set during the training process, and the generalization ability of the Hua Tuo model is evaluated based on the test set, thereby obtaining a trained improved Hua Tuo model.

[0065] You can set appropriate training parameters such as learning rate, batch size, and number of training rounds according to the model requirements and computing resources. The learning rate determines the step size of the model parameter update, the batch size affects the efficiency and stability of the model training, and the number of training rounds determines the number of iterations of the model training.

[0066] During the training process, the loss function of the model is calculated based on the output of the model and the annotated true results, and the parameters of the model are updated using an optimization algorithm to reduce the value of the loss function. In one embodiment, during the training process, the weight file of the model can be saved regularly so that the training can be resumed when the training is interrupted or a problem occurs.

[0067] In one embodiment, the trained model is evaluated using a validation set, and the accuracy, recall and other indicators of the model on the validation set are calculated to evaluate the performance of the model for entity extraction. Based on the evaluation results, the problems and deficiencies of the model are analyzed.

[0068] In one embodiment, the model can be optimized based on the results of the model evaluation. The training parameters can be adjusted, the amount of training data can be increased, the quality of data annotation can be improved, etc., to improve the performance of the model. If the model has overfitting or underfitting problems, regularization techniques, increasing data diversity, etc. can be used to solve them.

[0069] Step S2006: Use the trained improved Hua Tuo large model as the first entity object neural network recognition model.

[0070] The first entity object neural network recognition model obtained based on the above embodiment can be well embedded in the RPA process to extract entity objects and reduce development costs.

[0071] Step S300: Based on the third RPA process, extract the structured data in the electronic medical record system and store it in the information page database.

[0072] Step S400, based on the fourth RPA process, extract the unstructured data in the electronic medical record system, based on the extracted unstructured data in the electronic medical record system, use the trained second entity object neural network recognition model to extract the second entity object in the unstructured data, and store the extracted second entity object in the information page database.

[0073] In one embodiment, the training method of the second entity object neural network recognition model can refer to the training method of the first entity object neural network recognition model to achieve the extraction of the second entity object, which will not be repeated here.

[0074] Step S500, correspond, match and edit the data in the information page database with the data items in the structured information page to obtain the information page corresponding to each patient.

[0075] For the information page database and information page, some data can be directly filled in, such as the name item is corresponded and matched, and the name in the read information page database is directly filled in the information page. However, there are many data that cannot be filled in directly. In view of this, in one embodiment, step S500 includes: corresponding to the corresponding data item, and matching the corresponding data item options to edit and select the matched options. For example, for the gender data item, in the information page, male is 1 and female is 2. At this time, it is necessary to first correspond to the corresponding gender data item, and then match the corresponding data item options. For example, the matched data option is 1 for male gender, and then edit and select the matched option, that is, fill 1 into the option filling position of the gender data item.

[0076] In one embodiment, the generated information page is named and saved as the patient's name + clinic time, for example, "XXX+XX year X month X day".

[0077] In one embodiment, the generated information page is saved in an editable document format such as Word.

[0078] Based on the above-mentioned embodiment of generating information pages, on the one hand, it can be applied to multiple versions of systems, with better compatibility and low development cost. On the other hand, once the HIS system of the mainstream HIS manufacturer and the electronic medical record system of the electronic medical record manufacturer on the market have been accurately adapted once, then in the future, if other hospitals use the adapted HIS system and electronic medical record system, they do not need to be adapted again. Deploying a software robot that automatically generates information pages in this hospital can greatly save deployment time. Therefore, the information page generation method adopted in the embodiment of the present application has high market value and economic value.

[0079] In one embodiment of the present application, a computer-readable storage medium is provided, on which a program is stored. The stored program includes a method that can be loaded by a processor and process any of the above embodiments.

[0080] Those skilled in the art will appreciate that all or part of the functions of the various methods in the above-mentioned embodiments can be implemented by hardware or by computer programs. When all or part of the functions in the above-mentioned embodiments are implemented by computer programs, the program can be stored in a computer-readable storage medium, and the storage medium can include: read-only memory, random access memory, disk, optical disk, hard disk, etc., and the program is executed by a computer to implement the above-mentioned functions. For example, the program is stored in the memory of the device, and when the program in the memory is executed by the processor, all or part of the above-mentioned functions can be implemented. In addition, when all or part of the functions in the above-mentioned embodiments are implemented by computer programs, the program can also be stored in a storage medium such as a server, another computer, disk, optical disk, flash disk or mobile hard disk, and can be downloaded or copied and saved in the memory of the local device, or the system of the local device is updated, and when the program in the memory is executed by the processor, all or part of the functions in the above-mentioned embodiments can be implemented.

[0081] The above specific examples are used to illustrate the present invention, which is only used to help understand the present invention and is not intended to limit the present invention. For those skilled in the art, according to the concept of the present invention, some simple deductions, modifications or substitutions can be made.

Claims

1. A method for generating an outpatient and emergency department treatment information page, characterized in that: include: Based on the first RPA process, the structured data in the HIS system is extracted and stored in the information page database; Based on the second RPA process, unstructured data in the HIS system is extracted, and based on the extracted unstructured data in the HIS system, the trained first entity object neural network recognition model is used to extract the first entity object in the unstructured data, and the extracted first entity object is stored in the information page database; Based on the third RPA process, the structured data in the electronic medical record system is extracted and stored in the information page database; Based on the fourth RPA process, unstructured data in the electronic medical record system is extracted, and based on the extracted unstructured data in the electronic medical record system, the trained second entity object neural network recognition model is used to extract the second entity object in the unstructured data, and the extracted second entity object is stored in the information page database; The data in the information page database are corresponded, matched and edited with the data items in the structured information page to obtain the information page corresponding to each patient.

2. The method for generating an outpatient and emergency treatment information page according to claim 1, characterized in that: Design methods for the first RPA process and / or the third RPA process include: Determine the structured data that needs to be extracted as the target structured data; Simulate the operation of manually logging into the HIS system and / or electronic medical record system; Navigate to the page containing the target structured data through manual operation; Based on data crawling or reading functions, target structured data is extracted according to the preset location and format.

3. The method for generating an outpatient and emergency department treatment information page according to claim 2, characterized in that: The design method of the first RPA process and / or the third RPA process further includes: Perform data conversion and cleaning on the extracted data.

4. The method for generating an outpatient and emergency department treatment information page according to claim 1, characterized in that: The design method of the second RPA process and / or the fourth RPA process includes: Determine the specific business scenarios of the target unstructured data that needs to be extracted in the HIS system and / or electronic medical record system, and determine the type, source, and specific fields or information of the target unstructured data to be extracted; Simulate the operation of manually logging into the HIS system and / or electronic medical record system; Navigate to the page or document containing the target unstructured data through manual operation; According to the characteristics and extraction requirements of the target unstructured data, set corresponding data extraction rules to extract the target unstructured data.

5. The method for generating an outpatient and emergency department treatment information page according to claim 1, characterized in that: The training method of the first entity object neural network recognition model includes: Collecting a plurality of unstructured medical text data, wherein the medical text data includes a first entity object to be extracted; Annotating the medical text data with a first entity object that needs to be extracted; Convert the annotated medical text data into a format suitable for the Hua Tuo model input to obtain annotated data; Dividing the labeled data into a training set, a validation set, and a test set; Based on the preset training parameters, the training set is used as the input of the Hua Tuo model for training. The performance of the Hua Tuo model is evaluated and the hyperparameters are adjusted based on the validation set during the training process. The generalization ability of the Hua Tuo model is evaluated based on the test set, thereby obtaining a trained and improved Hua Tuo model. The trained improved Hua Tuo model is used as the first entity object neural network recognition model.

6. The method for generating an outpatient and emergency treatment information page according to claim 1, characterized in that: The training method of the second entity object neural network recognition model includes: Collecting a plurality of unstructured medical text data, wherein the medical text data includes a second entity object to be extracted; Annotating the medical text data with a second entity object that needs to be extracted; Convert the annotated medical text data into a format suitable for the Hua Tuo model input to obtain annotated data; Dividing the labeled data into a training set, a validation set, and a test set; Based on the preset training parameters, the training set is used as the input of the Hua Tuo model for training. The performance of the Hua Tuo model is evaluated and the hyperparameters are adjusted based on the validation set during the training process. The generalization ability of the Hua Tuo model is evaluated based on the test set, thereby obtaining a trained and improved Hua Tuo model. The trained improved Hua Tuo model is used as the second entity object neural network recognition model.

7. The method for generating an outpatient and emergency department treatment information page according to claim 1, characterized in that: The said corresponding, matching and editing of the data in the information page database with the data items in the structured information page includes: corresponding to the corresponding data items, and matching the corresponding data item options to edit and select the matched options.

8. The method for generating an outpatient and emergency treatment information page according to claim 1, characterized in that: Also includes: The generated information page is saved as patient name + clinic time.

9. An RPA process design method, characterized in that: Used to obtain the RPA process in the outpatient and emergency treatment information page generation method as described in any one of claims 1 to 8, including obtaining a first RPA process, a second RPA process, a third RPA process and a fourth RPA process; the design method includes: Obtaining the content that needs to be edited on the information page and classifying it, wherein the classification includes the content that needs to be extracted from the HIS system and the content that needs to be extracted from the electronic medical record system; Obtaining a classification of content that needs to be extracted from the HIS system, and further classifying the content into first structured data and first unstructured data; Based on the first structured data, a first RPA process is designed, and based on the first unstructured data, a second RPA process is designed; Obtaining a classification of content that needs to be extracted from the electronic medical record system, and further classifying the content into second structured data and second unstructured data; Based on the second structured data, a third RPA process is designed, and based on the fourth unstructured data, a fourth RPA process is designed.

10. A computer-readable storage medium, characterized in that: The medium stores a program, which can be loaded by a processor and execute the outpatient and emergency treatment information page generation method as described in any one of claims 1 to 8 and / or the RPA process design method as described in claim 9.

Images