Medical information determination method, device, equipment and medium

Abstract

The present invention provides a medical information determination method, device, equipment and storage medium, which can be applied to the medical technology field and the big data field. The method includes: performing text recognition on the acquired medical page information to obtain page text data, the page text data includes initial medical attribute description text and initial medical attribute data corresponding to the initial medical attribute description text; inputting the page text data into the medical data update model, outputting target text data, the target text data includes a target medical attribute name and target medical attribute data corresponding to the initial medical attribute description text, the target medical attribute data has a target data attribute associated with the target medical attribute name; and generating target medical information representing the user's medical condition according to the target medical attribute name and the target medical attribute data. The present invention achieves the technical effect of improving the integrity, accuracy and reliability of the user's medical information.

Description

Technical Field

[0001] The present invention relates to the fields of medical technology and big data, and in particular to a method, device, equipment and storage medium for determining medical information. Background Art

[0002] With the rapid development of science and technology, hospitals, rehabilitation institutions and other medical institutions can record and store information related to users' medical conditions by building medical service systems. For example, they can store medical data such as patients' medical examination results and diagnosis results to provide a reliable basis for patients' subsequent treatment and rehabilitation. However, it is difficult to unify the data standards of medical information of different medical institutions, which makes it difficult to share medical information between different medical institutions. It also takes a lot of time and manpower to analyze the medical information of different medical institutions, which has a negative impact on the quality or rehabilitation process of users. Summary of the invention

[0003] In view of the above problems, the present invention provides a medical information determination method, apparatus, device and storage medium.

[0004] According to a first aspect of the present invention, a medical information determination method is provided, characterized in that it includes: performing text recognition on acquired medical page information to obtain page text data, wherein the page text data includes initial medical attribute description text and initial medical attribute data corresponding to the initial medical attribute description text; inputting the page text data into a medical data update model to output target text data, wherein the target text data includes a target medical attribute name and target medical attribute data corresponding to the initial medical attribute description text, and the target medical attribute data has a target data attribute associated with the target medical attribute name; and generating target medical information representing the medical condition of the user based on the target medical attribute name and the target medical attribute data.

[0005] The second aspect of the present invention provides a medical information determination device, characterized in that it includes: a text recognition module, which is used to perform text recognition on the acquired medical page information to obtain page text data, wherein the above-mentioned page text data includes an initial medical attribute description text and initial medical attribute data corresponding to the above-mentioned initial medical attribute description text; a target text data acquisition module, which is used to input the above-mentioned page text data into a medical data update model and output target text data, wherein the above-mentioned target text data includes a target medical attribute name and target medical attribute data corresponding to the above-mentioned initial medical attribute description text, and the above-mentioned target medical attribute data has a target data attribute associated with the above-mentioned target medical attribute name; and a target medical information acquisition module, which is used to generate target medical information representing the medical condition of the above-mentioned user based on the above-mentioned target medical attribute name and the above-mentioned target medical attribute data.

[0006] A third aspect of the present invention provides an electronic device, comprising: one or more processors; a memory for storing one or more programs, wherein when the one or more programs are executed by the one or more processors, the one or more processors execute the above method.

[0007] The fourth aspect of the present invention also provides a computer-readable storage medium having executable instructions stored thereon, which, when executed by a processor, causes the processor to execute the above method.

[0008] The fifth aspect of the present invention also provides a computer program product, including a computer program, which implements the above method when executed by a processor.

[0009] According to the medical information determination method, device, equipment and medium provided by the present invention, by inputting the page text data identified from the medical page information into the medical text prediction model, the initial medical attribute description text can be updated to a standardized target medical attribute name, and the initial medical attribute data can be updated to a target medical attribute data with target data attributes, so that standardized target medical information can be obtained according to the target medical attribute name and the target medical attribute data, at least partially avoiding technical problems such as large differences in extracted medical attribute names and ambiguous names, and ensuring the consistency of data attributes of medical attribute data, thereby achieving the technical effect of improving the integrity, accuracy and reliability of the user's medical information, and facilitating relevant personnel to conduct medical services such as medical condition inquiries, medical testing, and diagnosis. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] The above contents and other objects, features and advantages of the present invention will become more apparent through the following description of the embodiments of the present invention with reference to the accompanying drawings, in which:

[0011] Figure 1 An application scenario diagram of a medical information determination method and device according to an embodiment of the present invention is shown;

[0012] Figure 2 A flowchart of a method for determining medical information according to an embodiment of the present invention is shown;

[0013] Figure 3 A flow chart showing a training method of a medical data updating model according to an embodiment of the present invention is shown;

[0014] Figure 4 A flowchart of deploying a medical data update model according to an embodiment of the present invention is shown;

[0015] Figure 5 It shows a structural block diagram of a medical information determination device according to an embodiment of the present invention;

[0016] Figure 6 A block diagram of an electronic device suitable for implementing a medical information determination method according to an embodiment of the present invention is shown. DETAILED DESCRIPTION

[0017] Below, embodiments of the present invention will be described with reference to the accompanying drawings. However, it should be understood that these descriptions are exemplary only and are not intended to limit the scope of the present invention. In the following detailed description, for ease of explanation, many specific details are set forth to provide a comprehensive understanding of embodiments of the present invention. However, it is apparent that one or more embodiments may also be implemented without these specific details. In addition, in the following description, descriptions of known structures and technologies are omitted to avoid unnecessary confusion of concepts of the present invention.

[0018] The terms used herein are only for describing specific embodiments and are not intended to limit the present invention. The terms "comprise", "include", etc. used herein indicate the existence of the features, steps, operations and / or components, but do not exclude the existence or addition of one or more other features, steps, operations or components.

[0019] All terms (including technical and scientific terms) used herein have the meanings commonly understood by those skilled in the art, unless otherwise defined. It should be noted that the terms used herein should be interpreted as having a meaning consistent with the context of this specification, and should not be interpreted in an idealized or overly rigid manner.

[0020] When using expressions such as "at least one of A, B, and C, etc.", they should generally be interpreted according to the meaning of the expression commonly understood by those skilled in the art (for example, "a system having at least one of A, B, and C" should include but is not limited to a system having A alone, B alone, C alone, A and B, A and C, B and C, and / or A, B, C, etc.).

[0021] In the technical solution of the present invention, the collection, storage, use, processing, transmission, provision, disclosure and application of user personal information involved comply with the provisions of relevant laws and regulations, take necessary confidentiality measures, and do not violate public order and good morals.

[0022] In the technical solution of the present invention, the acquisition, collection, storage, use, processing, transmission, provision, disclosure and application of data all comply with the provisions of relevant laws and regulations, take necessary confidentiality measures, and do not violate public order and good morals.

[0023] The inventors have discovered that the standards for medical information between different medical service systems (or medical systems) are becoming more diversified, which causes great trouble to medical research and patient medical diagnosis. In addition, conventional medical information collection methods are difficult to accurately characterize the medical attributes of medical information, and the data collection method process is relatively long, making it difficult to accurately match standard medical information, resulting in a high error rate in the collected medical information, making it difficult to apply to actual medical business application scenarios such as medical research and medical diagnosis.

[0024] An embodiment of the present invention provides a medical information determination method, comprising: performing text recognition on acquired medical page information to obtain page text data, wherein the page text data includes initial medical attribute description text and initial medical attribute data corresponding to the initial medical attribute description text; inputting the page text data into a medical data update model to output target text data, wherein the target text data includes a target medical attribute name and target medical attribute data corresponding to the initial medical attribute description text, and the target medical attribute data has a target data attribute associated with the target medical attribute name; and generating target medical information characterizing the user's medical condition based on the target medical attribute name and the target medical attribute data.

[0025] According to an embodiment of the present invention, by inputting page text data identified from medical page information into a medical text prediction model, it is possible to update the initial medical attribute description text to a standardized target medical attribute name, and it is also possible to update the initial medical attribute data to target medical attribute data having target data attributes, so that standardized target medical information can be obtained based on the target medical attribute name and the target medical attribute data, at least partially avoiding technical problems such as large differences in extracted medical attribute names and ambiguous names, and ensuring the consistency of data attributes of medical attribute data, thereby achieving the technical effect of improving the integrity, accuracy and reliability of the user's medical information, and facilitating relevant personnel to conduct medical services such as medical condition inquiries, medical testing, and diagnosis.

[0026] Figure 1 An application scenario diagram of a medical information determination method and device according to an embodiment of the present invention is shown.

[0027] like Figure 1 As shown, the application scenario 100 according to this embodiment may include a first terminal device 101, a second terminal device 102, a third terminal device 103, a network 104, and a server 105. The network 104 is used to provide a medium for a communication link between the first terminal device 101, the second terminal device 102, the third terminal device 103, and the server 105. The network 104 may include various connection types, such as wired, wireless communication links, or optical fiber cables, etc.

[0028] The user can use the first terminal device 101, the second terminal device 102, and the third terminal device 103 to interact with the server 105 through the network 104 to receive or send messages, etc. Various communication client applications can be installed on the first terminal device 101, the second terminal device 102, and the third terminal device 103, such as shopping applications, web browser applications, search applications, instant messaging tools, email clients, social platform software, etc. (only for example).

[0029] The first terminal device 101, the second terminal device 102, and the third terminal device 103 may be various electronic devices having display screens and supporting web browsing, including but not limited to smart phones, tablet computers, laptop computers, desktop computers, and the like.

[0030] The server 105 may be a server that provides various services, such as a background management server (only as an example) that provides support for websites browsed by users using the first terminal device 101, the second terminal device 102, and the third terminal device 103. The background management server may analyze and process the received data such as user requests, and feed back the processing results (such as web pages, information, or data obtained or generated according to user requests) to the terminal device.

[0031] It should be noted that the medical information determination method provided in the embodiment of the present invention can generally be executed by the server 105. Accordingly, the medical information determination device provided in the embodiment of the present invention can generally be set in the server 105. The medical information determination method provided in the embodiment of the present invention can also be executed by a server or server cluster that is different from the server 105 and can communicate with the first terminal device 101, the second terminal device 102, the third terminal device 103 and / or the server 105. Correspondingly, the medical information determination device provided in the embodiment of the present invention can also be set in a server or server cluster that is different from the server 105 and can communicate with the first terminal device 101, the second terminal device 102, the third terminal device 103 and / or the server 105.

[0032] It should be understood that Figure 1 The number of terminal devices, networks and servers in the embodiment is only for illustration. Any number of terminal devices, networks and servers may be provided according to the implementation requirements.

[0033] The following will be based on Figure 1 The scene described by Figure 2 to Figure 4 The medical information determination method of the disclosed embodiment is described in detail.

[0034] Figure 2 A flowchart of a medical information determination method according to an embodiment of the present invention is shown.

[0035] like Figure 2 As shown, the medical information determination method of this embodiment includes operations S210 to S230.

[0036] In operation S210, text recognition is performed on the acquired medical page information to obtain page text data.

[0037] According to an embodiment of the present invention, medical page information may include a page recording medical information of a medical subject (e.g., a patient, a physical examination personnel), and the medical page information may be, for example, a patient's medical record page, a medical examination page, a medical imaging report page, etc. The embodiment of the present invention does not limit the specific type of medical page information, and those skilled in the art may select it according to actual needs.

[0038] According to an embodiment of the present invention, page text data can be any type of text data such as text, numerical value, attribute unit identifier, etc. extracted from medical page information. The embodiment of the present invention does not limit the specific type of page text data, and technical personnel in this field can choose according to actual needs.

[0039] According to an embodiment of the present invention, page text data may include initial medical attribute description text and initial medical attribute data corresponding to the initial medical attribute description text. The initial medical attribute description text may be text used to describe the medical symptoms of a medical object. For example, the initial medical attribute description text may be text used to describe medical symptoms such as "fever", "fever", "above 37.5°C", etc. Alternatively, the initial medical attribute description text may also be text used to describe medical indicators. For example, the initial medical attribute description text may be "height", "height", etc. The initial medical attribute data may include indicator data corresponding to the initial medical attribute description text. For example, the initial medical attribute data corresponding to "height" may be "1800 mm", "1.8 meters", etc.

[0040] It should be noted that the page text data may be an array including initial medical attribute description text and initial medical attribute data, or may be a description text including initial medical attribute description text and initial medical attribute data, such as a diagnosis certificate text.

[0041] In operation S220, the page text data is input into the medical data update model, and target text data is output, wherein the target text data includes a target medical attribute name and target medical attribute data corresponding to the initial medical attribute description text, and the target medical attribute data has a target data attribute associated with the target medical attribute name.

[0042] According to an embodiment of the present invention, the medical data update model can be a model constructed based on a deep learning algorithm, for example, it can be constructed based on an attention network algorithm, or it can also be constructed based on other types of deep learning algorithms. The embodiment of the present invention does not limit the specific algorithm type for constructing the medical data update model, and technical personnel in this field can make a choice based on actual needs.

[0043] According to an embodiment of the present invention, the target medical attribute name corresponding to the initial medical attribute description text may include a standardized, unified medical symptom name, for example, the target medical attribute name corresponding to "above 37.5°C" (initial medical attribute description text) may be "fever". By obtaining the target medical attribute name, semantic information related to the medical attribute name in the initial medical attribute description text may be identified, and the semantic information may be standardized to obtain standardized target medical attribute names such as medical symptoms and medical indicators.

[0044] According to an embodiment of the present invention, the target medical attribute data may be obtained by updating the data attributes of the initial medical attribute data. For example, the target medical attribute data "180 cm" corresponding to the initial medical attribute data "1800 mm" may be obtained.

[0045] According to an embodiment of the present invention, the target data attribute may be a standardized data attribute corresponding to the target medical attribute name, and the target data attribute may include a data value range, a data unit (e.g., centimeters), a data length, etc. The embodiment of the present invention does not limit the specific attribute type of the target data attribute, as long as it can correspond to the target quality attribute indicator name.

[0046] According to an embodiment of the present invention, the target text data may be structured data such as an array or a key-value pair including the target medical attribute name and the target medical attribute data, or may also be a descriptive text including the target medical attribute name and the target medical attribute data. The embodiment of the present invention does not limit the specific type of the target text data, and those skilled in the art may make a selection based on actual needs.

[0047] In operation S230, target medical information representing the medical condition of the user is generated according to the target medical attribute name and the target medical attribute data.

[0048] According to the embodiment of the present invention, the target medical information may be any type of information such as text, table, etc. The embodiment of the present invention does not limit the specific information type of the target medical information.

[0049] For example, a standardized diagnosis and treatment report can be generated based on the target medical attribute name and target medical attribute data, so that subsequent users or relevant medical staff can understand the user's medical condition by conveniently viewing the diagnosis and treatment report, thereby improving the treatment level for the user. Alternatively, the target medical information can also be processed by a related automated detection system, so that relevant medical auxiliary suggestions can be quickly generated by processing standardized target medical attribute names and target medical attribute data, thereby improving the calculation speed and prediction accuracy of the automated detection system.

[0050] According to an embodiment of the present invention, the initial medical attribute description text may include at least one of the following: physical sign indicator description data, symptom indicator description data, and physical examination indicator data.

[0051] According to an embodiment of the present invention, the physical sign index description data may be, for example, index data related to physical signs such as gender, age, height, weight, blood pressure, blood sugar, and body temperature. The physical examination index data may be, for example, index data related to physical examination results or test results such as blood routine index, urine routine index, liver function index, kidney function index, chest X-ray, etc. The disease index description data may be, for example, data related to disease indicators such as chronic obstructive pulmonary disease index description data, colorectal cancer index description data, stroke index description data, and myocardial infarction index description data.

[0052] According to an embodiment of the present invention, the initial medical attribute description text may also include any type of medical-related data such as stool index description data, urine test index description data, chest X-ray index description data, etc. The embodiment of the present disclosure does not limit the specific data type of the initial medical attribute data.

[0053] According to an embodiment of the present invention, the body temperature index description data may be, for example, above 37.5 degrees, and the blood index description data may be, for example, "the blood oxygen concentration is below 94%".

[0054] According to an embodiment of the present invention, the target data attribute includes at least one of the following: a data type, a data value threshold, and a data value unit identifier.

[0055] According to an embodiment of the present invention, the data type may include any type such as a numerical type, a text type, etc. The data value threshold may represent a data value range related to the target medical attribute data, and the data unit identifier may indicate a data unit corresponding to the target medical attribute name, for example, indicating that the data unit corresponding to "height" is "centimeter".

[0056] According to an embodiment of the present invention, performing text recognition on the acquired medical page information to obtain page text data may include: performing image capture on the acquired medical page information to obtain a medical page image, wherein the medical page information records medical data related to the user; and performing text recognition on the medical page image to obtain page text data.

[0057] According to an embodiment of the present invention, the system page of the hospital clinical business system can be automatically scrolled and captured to obtain a medical page image. The medical page image can be subjected to text recognition using OCR (Optical Character Recognition) technology to extract text information from the medical page image to obtain page text data. The page text data can be a description text obtained after splicing that includes the initial medical attribute description text and the initial medical attribute data. For example, a scrolling screenshot can be taken of a multi-page medical record report, and the complete content of each web page can be identified from the screenshot.

[0058] In one embodiment of the present invention, in order to ensure the integrity of the user's case information, page text data may also be generated based on the following data preprocessing process.

[0059] The medical page is screenshotted in a rolling screenshot mode to obtain continuous page screenshots. The continuous page screenshots can have a certain amount of overlapping page content, so as to ensure that the page information obtained after OCR recognition can completely retain the cases of related user objects through the data post-processing strategy. The specific data post-processing strategy includes the following steps:

[0060] 1) Content redundancy removal: Perform OCR recognition on consecutive page screenshots in the order of screenshots to obtain continuous page text information. Perform preliminary redundant content removal based on the overlap between the previous page content and the subsequent page content to ensure that the case information is coherent and complete.

[0061] 2) Terminology Correction: Medical language models (such as N-gram language models or language models built based on neural networks) are trained by publicly available outdoor medical research data. The medical language models are used to correct text content after redundant content is removed, correct obvious text recognition errors, and ensure that the text after redundant content is removed is aligned with medical text terminology.

[0062] 3) Splicing: Splice the text obtained in step 2) in the order of the screenshots to obtain complete user case information for prediction by the medical data update model. The spliced ​​text obtained in step 3) can also be used to train the medical data update model.

[0063] According to the embodiment of the present invention, compared with directly extracting text information from medical page information to obtain page text data, the method provided by the embodiment of the present invention can be used to obtain page text data, which can avoid developing corresponding text extraction scripts for different medical page information, thereby improving the efficiency of obtaining page text data. At the same time, based on text recognition of medical page images, page text data corresponding to the distribution format of medical page information can be obtained, and the format attributes of the initial medical attribute description text and the initial medical attribute data in the page text data have been retained.

[0064] According to an embodiment of the present invention, performing text recognition on a medical page image to obtain page text data includes: identifying text data in the medical page image to obtain initial page text data; performing word segmentation processing on the initial page text data based on preset word segmentation rules to obtain page text data containing initial medical attribute description text and initial medical attribute data.

[0065] According to the embodiment of the present invention, the text content in the medical page image can be identified based on OCR technology, and the text content can be converted into a digital text format that can be retrieved and edited. At the same time, the initial page text data is segmented by the preset segmentation rules obtained by constructing a standardized medical field dictionary to obtain page text data containing initial medical attribute description text and initial medical attribute data.

[0066] According to an embodiment of the present invention, a standardized standard medical field dictionary can be constructed based on medical-related industry data standard information, so that word segmentation can be performed according to medical industry standards, so that subsequent medical data update models can more easily understand the semantic information in page text data and improve the prediction accuracy of target text data.

[0067] According to an embodiment of the present invention, the medical data update model can be constructed based on an AIGC (AI-Generated Content) model. For example, it can be a text prediction model constructed based on an adversarial network algorithm. By inputting the page text data into the AIGC model obtained after training, based on the powerful semantic understanding and text prediction capabilities of the AIGC model, the prediction accuracy of the target medical attribute name and the target medical attribute data can be improved, and a key-value pair containing the target medical attribute name and the target medical attribute data can be generated, so as to generate standardized and structured medical information based on the medical page information, thereby improving the accuracy of the constructed target medical information.

[0068] In another embodiment of the present disclosure, the medical data update model can be constructed based on a large language model (LLM) that is fine-tuned and trained.

[0069] It should be noted that the medical data update model can be deployed on a local server, or can also be deployed on a cloud server, so as to provide unified standardized services to various medical service systems through an API interface.

[0070] According to an embodiment of the present invention, when the target text data is obtained, further processing can be performed on the custom content in the target text data or the data that is difficult to predict by the medical data update model. For example, the difficult-to-predict data can be further standardized based on the standard template and configuration rules, and negative cases can be generated to adjust the parameters of the medical data update model and improve the prediction accuracy of the medical data update model. The generated target medical attribute name and target medical attribute data can be automatically stored in the standard database and backed up. Correspondingly, the target medical attribute name and target medical attribute data involving information security and personal privacy can also be encrypted or desensitized to improve the level of information security.

[0071] Figure 3 A flow chart of a training method for a medical data updating model according to an embodiment of the present invention is shown.

[0072] like Figure 3 As shown, the medical data update model can be trained based on the following training method. The training method of this embodiment includes operations S310 to S330.

[0073] In operation S310 , a training sample is acquired, wherein the training sample includes sample medical attribute description data and sample medical attribute data, as well as a label medical attribute name and label medical attribute data.

[0074] In operation S320, the sample medical attribute description data and the sample medical attribute data are input into the initial medical data update model, and the sample target medical attribute name and the sample target medical attribute data are output.

[0075] In operation S330, an initial medical data update model is trained according to the sample target medical attribute name and the sample target medical attribute data, and the label medical attribute name and the label medical attribute data to obtain a medical data update model.

[0076] According to an embodiment of the present invention, the sample target medical attribute name and the label medical attribute name can be input into a first loss function to obtain a first loss value, and the sample target medical attribute data and the label medical attribute data can be input into a second loss function to obtain a second loss value. The model parameters of the initial medical data update model are adjusted based on the first loss value and the second loss value until the first loss value and the second loss value converge, thereby obtaining a trained medical data update model.

[0077] It should be noted that the technical terms involved in the training method provided in the embodiment of the present invention, including but not limited to sample medical attribute description data, sample medical attribute data, etc., have the same or corresponding technical attributes as the technical terms involved in the above-mentioned embodiments of the present invention, including but not limited to initial medical attribute description text, initial medical attribute data, etc., and the embodiments of the present invention will not be repeated here.

[0078] According to an embodiment of the present invention, obtaining training samples may include: performing text recognition on the obtained sample medical page image to obtain sample medical attribute description data and sample medical attribute data; determining the label medical attribute name associated with the sample medical attribute description data from a preset standard medical attribute data table; and updating the sample medical attribute data according to the indicator data attributes associated with the label medical attribute name to obtain the label medical attribute data.

[0079] According to an embodiment of the present invention, based on relevant industry data standard information and according to medical business standards, standardized target medical attribute names can be extracted from industry data standard information to form a standardized standard dictionary. Accordingly, a standard medical attribute data table can also be constructed by constructing a mapping relationship between sample medical attribute description data and target medical attribute names in a standardized standard dictionary, as well as a mapping relationship between target data attributes of target medical attribute data and target medical attribute names.

[0080] For example, based on expert experience, a standard medical attribute data table can be constructed through the following steps. It is mainly divided into the following steps:

[0081] Medical core data elements are determined: Medical core data elements (initial medical attribute description text) can include basic patient information, diagnosis information, treatment information, laboratory test results, imaging test results, etc. The mapping relationship between medical core data elements and target medical attribute names is established by uniformly defining medical core data elements and naming them based on target medical attribute names.

[0082] Data classification and grading: A large number of core medical data elements can be classified and graded to facilitate data management and query. For example, basic patient information, diagnosis information, treatment information, examination results, etc. can be classified according to their respective characteristics, and then organized and stored according to the hierarchical structure of the characteristic classification.

[0083] Attribute description of data elements: Each core medical data element can correspond to a detailed data attribute description, which can include data type, value range, data value unit identifier, etc. By establishing a mapping relationship between the target data attribute and the target medical attribute name, each medical system can correctly understand the meaning of the target medical attribute data.

[0084] According to an embodiment of the present invention, the training method may further include: performing a desensitization operation on the training samples to obtain the desensitized training samples.

[0085] According to an embodiment of the present invention, the sample medical attribute description data and sample medical attribute data, as well as the label medical attribute name and label medical attribute data in the training sample can also be desensitized to ensure the integrity and security of the training sample. Accordingly, the corresponding data in the standard medical attribute data table can also be desensitized to protect the patient's personal privacy.

[0086] According to an embodiment of the present invention, during the application of the medical information determination method, the patient's privacy data can also be desensitized, and during the data exchange with the medical service system, corresponding security policies and security technical standards can be followed to avoid privacy leakage.

[0087] According to an embodiment of the present invention, the acquired sample medical indicator description data and sample medical indicator data may also be labeled in the following manner to obtain a label medical indicator name and label medical indicator data.

[0088] A small amount of sample data can be collected for different medical service systems as input data for the subsequent initial medical data update model. Accordingly, the sample medical indicator description data and sample medical indicator data obtained by OCR technology can be labeled through the standard medical attribute data table, so as to obtain a key-value pair consisting of the labeled medical attribute name and the labeled medical attribute data.

[0089] When designing the annotation system, in order to lower the threshold for annotation personnel and improve the convenience of operation, the annotation system can provide an OCR visual text interface, and the annotated user can intuitively view the original sample medical indicator description data and sample medical indicator data recognized by OCR. At the same time, the annotation system also provides a fast alignment function with the standardized standard medical attribute data table, allowing users to quickly map the recognized label information to the corresponding standard medical attribute data table.

[0090] The training process can be performed based on GPU (Graphics Processing Unit) hardware resources to meet the hardware requirements for deep learning model training.

[0091] Each piece of annotated data generated by the annotation system, namely the sample medical indicator description data and sample medical indicator data recognized by OCR, as well as the labeled medical attribute name and labeled medical attribute data obtained after annotation, is input into the initial medical data update model. After optimizing each model parameter, the trained medical data update model can be obtained.

[0092] According to an embodiment of the present invention, the trained medical data update model can be evaluated based on the accuracy and recall rate, and the classification of inaccurately identified medical data can be retrained until the test requirements are met.

[0093] Figure 4 A flowchart of deploying a medical data update model according to an embodiment of the present invention is shown.

[0094] like Figure 4 As shown, the standardization of medical data can be achieved by first constructing a standard medical attribute data table. Then, sample medical attribute description data and sample medical attribute data are extracted, and the sample medical attribute description data and sample medical attribute data are labeled according to the standard medical attribute data table in the labeling system to obtain label medical attribute names and label medical attribute data. Then, a medical data update model is obtained based on the labeled label medical attribute names and label medical attribute data, as well as the sample medical attribute description data and sample medical attribute data. By evaluating and fine-tuning the medical data update model, a medical data update model that can be deployed on the cloud or local server is obtained.

[0095] According to an embodiment of the present invention, the medical information determination method may further include: sending a login request to the medical system; and upon receiving a login success message sent by the medical system, acquiring medical page information from the medical system.

[0096] According to an embodiment of the present invention, a system deployed with a medical update model can provide a standard service interface to the medical system, and the hospital management personnel provide the system with the administrator account and password of the medical system, so that the system deployed with the medical update model can retrieve information and extract standardized page data from the medical system to form a standardized database. There is no need for the medical system to provide a data dictionary, data interface, etc. to cooperate, and standardized target medical information can be provided to the medical system.

[0097] According to an embodiment of the present invention, the medical update model can also be deployed on a cloud server, and provide service resources such as mini-programs, mobile clients, and web pages based on the cloud server. Users can log in to the mini-program or client, and send photos of paper medical examination reports and screenshots of medical examination pages provided by various medical service programs to the cloud server. The cloud server can generate standardized target medical information based on the medical update model, thereby realizing the construction of a standardized and unified electronic health record for users.

[0098] The technical solution adopted in the present invention aims at the problem of insufficient universality of medical business data collection. It uses the user's personal front-end authority or the front-end administrator's authority to obtain business data information through mobile terminal screenshots, uses OCR technology to identify the structured page text data in the screenshots, and processes the screenshot sequence and the screenshot text intersection information to complete the complete collection of medical attribute description text and medical attribute data, and is not limited to the front-end display style of medical information in different hospitals, thereby achieving the purpose of universal medical data collection. At the same time, the present invention can replace the medical field mapping and field value mapping based on the AIGC model obtained after training, directly use the powerful semantic understanding and reasoning ability of the AIGC model, and directly standardize the output of universal medical business metadata (initial medical attribute description text and initial medical attribute data corresponding to the initial medical attribute description text) to obtain standardized target medical attribute name and target medical attribute data, thereby avoiding the problem of cumulative propagation of multi-level process errors and greatly improving the accuracy of data collection. Finally, the present invention is not limited to specific medical business data storage or front-end display form from medical business metadata collection to data standardization, and truly achieves the purpose of universal collection and standardization of medical business data.

[0099] Based on the above-mentioned medical information determination method, the present invention also provides a medical information determination device. Figure 5 The device is described in detail.

[0100] Figure 5 A structural block diagram of a medical information determination device according to an embodiment of the present invention is shown.

[0101] like Figure 5 As shown, the medical information determination device 500 of this embodiment includes a text recognition module 510 , a target text data acquisition module 520 and a target medical information acquisition module 530 .

[0102] The text recognition module 510 is used to perform text recognition on the acquired medical page information to obtain page text data, wherein the page text data includes initial medical attribute description text and initial medical attribute data corresponding to the initial medical attribute description text.

[0103] The target text data acquisition module 520 is used to input the page text data into the medical data update model and output the target text data, wherein the target text data includes a target medical attribute name and target medical attribute data corresponding to the initial medical attribute description text, and the target medical attribute data has a target data attribute associated with the target medical attribute name.

[0104] The target medical information obtaining module 530 is used to generate target medical information representing the medical condition of the user according to the target medical attribute name and the target medical attribute data.

[0105] According to an embodiment of the present invention, the medical data update model is trained based on the following training method, which includes: obtaining training samples, wherein the training samples include sample medical attribute description data and sample medical attribute data, as well as label medical attribute names and label medical attribute data; inputting the sample medical attribute description data and the sample medical attribute data into the initial medical data update model, and outputting the sample target medical attribute name and the sample target medical attribute data; training the initial medical data update model according to the sample target medical attribute name and the sample target medical attribute data, as well as the label medical attribute name and the label medical attribute data, to obtain the medical data update model.

[0106] According to an embodiment of the present invention, obtaining training samples includes: performing text recognition on the obtained sample medical page image to obtain sample medical attribute description data and sample medical attribute data; determining the label medical attribute name associated with the sample medical attribute description data from a preset standard medical attribute data table; and updating the sample medical attribute data according to the indicator data attributes associated with the label medical attribute name to obtain the label medical attribute data.

[0107] According to an embodiment of the present invention, the training method further includes: performing a desensitization operation on the training samples to obtain the desensitized training samples.

[0108] According to an embodiment of the present invention, the target data attribute includes at least one of the following: a data type, a data value threshold, and a data value unit identifier.

[0109] According to an embodiment of the present invention, the text recognition module includes: an image acquisition unit and a text recognition unit.

[0110] The image acquisition unit is used to perform image acquisition on the acquired medical page information to obtain a medical page image, wherein the medical page information records medical data related to the user.

[0111] The text recognition unit is used to perform text recognition on the medical page image to obtain page text data.

[0112] According to an embodiment of the present invention, the text recognition unit includes: an initial page text data obtaining subunit and a page text data obtaining subunit.

[0113] The initial page text data obtaining subunit is used to identify the text data in the medical page image and obtain the initial page text data.

[0114] The page text data obtaining subunit is used to perform word segmentation processing on the initial page text data based on preset word segmentation rules to obtain page text data including initial medical attribute description text and initial medical attribute data.

[0115] According to an embodiment of the present invention, the initial medical attribute description text includes at least one of the following: physical sign indicator description data, physical examination indicator description data, and symptom indicator description data.

[0116] According to an embodiment of the present invention, the medical information determination device further includes: a sending module and a medical page information obtaining module.

[0117] The sending module is used to send a login request to the medical system.

[0118] The medical page information acquisition module is used to obtain medical page information from the medical system when a login success message sent by the medical system is received.

[0119] According to an embodiment of the present invention, any multiple modules of the text recognition module 510, the target text data acquisition module 520 and the target medical information acquisition module 530 can be combined into one module for implementation, or any one of the modules can be split into multiple modules. Alternatively, at least part of the functions of one or more of these modules can be combined with at least part of the functions of other modules and implemented in one module. According to an embodiment of the present invention, at least one of the text recognition module 510, the target text data acquisition module 520 and the target medical information acquisition module 530 can be at least partially implemented as a hardware circuit, such as a field programmable gate array (FPGA), a programmable logic array (PLA), a system on a chip, a system on a substrate, a system on a package, an application specific integrated circuit (ASIC), or can be implemented by hardware or firmware such as any other reasonable way of integrating or packaging the circuit, or implemented in any one of the three implementation methods of software, hardware and firmware or in any appropriate combination of any of them. Alternatively, at least one of the text recognition module 510, the target text data acquisition module 520 and the target medical information acquisition module 530 can be at least partially implemented as a computer program module, and when the computer program module is run, the corresponding function can be performed.

[0120] Figure 6 A block diagram of an electronic device suitable for implementing a medical information determination method according to an embodiment of the present invention is shown.

[0121] like Figure 6 As shown, the electronic device 600 according to an embodiment of the present invention includes a processor 601, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 602 or a program loaded from a storage part 608 into a random access memory (RAM) 603. The processor 601 may include, for example, a general-purpose microprocessor (e.g., a CPU), an instruction set processor and / or a related chipset and / or a special-purpose microprocessor (e.g., an application-specific integrated circuit (ASIC)), etc. The processor 601 may also include an onboard memory for caching purposes. The processor 601 may include a single processing unit or multiple processing units for performing different actions of the method flow according to an embodiment of the present invention.

[0122] In RAM 603, various programs and data required for the operation of electronic device 600 are stored. Processor 601, ROM 602 and RAM 603 are connected to each other via bus 604. Processor 601 performs various operations of the method flow according to the embodiment of the present invention by executing the program in ROM 602 and / or RAM 603. It should be noted that the program can also be stored in one or more memories other than ROM 602 and RAM 603. Processor 601 can also perform various operations of the method flow according to the embodiment of the present invention by executing the program stored in the one or more memories.

[0123] According to an embodiment of the present invention, the electronic device 600 may further include an input / output (I / O) interface 605, which is also connected to the bus 604. The electronic device 600 may further include one or more of the following components connected to the I / O interface 605: an input portion 606 including a keyboard, a mouse, etc.; an output portion 607 including a cathode ray tube (CRT), a liquid crystal display (LCD), etc., and a speaker, etc.; a storage portion 608 including a hard disk, etc.; and a communication portion 609 including a network interface card such as a LAN card, a modem, etc. The communication portion 609 performs communication processing via a network such as the Internet. A drive 610 is also connected to the I / O interface 605 as needed. A removable medium 611, such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc., is installed on the drive 610 as needed, so that a computer program read therefrom is installed into the storage portion 608 as needed.

[0124] The present invention also provides a computer-readable storage medium, which may be included in the device / apparatus / system described in the above embodiment; or may exist independently without being assembled into the device / apparatus / system. The above computer-readable storage medium carries one or more programs, and when the above one or more programs are executed, the method according to the embodiment of the present invention is implemented.

[0125] According to an embodiment of the present invention, the computer-readable storage medium may be a non-volatile computer-readable storage medium, for example, may include but is not limited to: a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination thereof. In the present invention, the computer-readable storage medium may be any tangible medium containing or storing a program, which may be used by or in combination with an instruction execution system, an apparatus or a device. For example, according to an embodiment of the present invention, the computer-readable storage medium may include the ROM 602 and / or RAM 603 described above and / or one or more memories other than ROM 602 and RAM 603.

[0126] The embodiment of the present invention also includes a computer program product, which includes a computer program, and the computer program contains program code for executing the method shown in the flowchart. When the computer program product is run in a computer system, the program code is used to enable the computer system to implement the method provided by the embodiment of the present invention.

[0127] The computer program executes the above functions defined in the system / device of the embodiment of the present invention when it is executed by the processor 601. According to the embodiment of the present invention, the system, device, module, unit, etc. described above can be implemented by a computer program module.

[0128] In one embodiment, the computer program may rely on tangible storage media such as optical storage devices, magnetic storage devices, etc. In another embodiment, the computer program may also be transmitted and distributed in the form of signals on a network medium, and downloaded and installed through the communication part 609, and / or installed from a removable medium 611. The program code contained in the computer program may be transmitted using any appropriate network medium, including but not limited to: wireless, wired, etc., or any suitable combination of the above.

[0129] In such an embodiment, the computer program can be downloaded and installed from the network through the communication part 609, and / or installed from the removable medium 611. When the computer program is executed by the processor 601, the above functions defined in the system of the embodiment of the present invention are performed. According to the embodiment of the present invention, the system, device, means, module, unit, etc. described above can be implemented by a computer program module.

[0130] According to an embodiment of the present invention, the program code for executing the computer program provided by the embodiment of the present invention can be written in any combination of one or more programming languages, specifically, these computing programs can be implemented using high-level process and / or object-oriented programming languages, and / or assembly / machine languages. Programming languages ​​include, but are not limited to, such as Java, C++, python, "C" language or similar programming languages. The program code can be executed completely on the user computing device, partially on the user device, partially on the remote computing device, or completely on the remote computing device or server. In the case of a remote computing device, the remote computing device can be connected to the user computing device through any type of network, including a local area network (LAN) or a wide area network (WAN), or can be connected to an external computing device (e.g., using an Internet service provider to connect through the Internet).

[0131] The flow chart and block diagram in the accompanying drawings illustrate the possible architecture, function and operation of the system, method and computer program product according to various embodiments of the present invention. In this regard, each box in the flow chart or block diagram can represent a module, a program segment, or a part of a code, and the above-mentioned module, program segment, or a part of a code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order from the order marked in the accompanying drawings. For example, two boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram or flow chart, and the combination of the boxes in the block diagram or flow chart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0132] It will be appreciated by those skilled in the art that the features described in the various embodiments and / or claims of the present invention may be combined and / or combined in various ways, even if such combinations and / or combinations are not explicitly described in the present invention. In particular, the features described in the various embodiments and / or claims of the present invention may be combined and / or combined in various ways without departing from the spirit and teachings of the present invention. All of these combinations and / or combinations fall within the scope of the present invention.

[0133] The embodiments of the present invention are described above. However, these embodiments are only for the purpose of illustration, and are not intended to limit the scope of the present invention. Although each embodiment is described above, it does not mean that the measures in each embodiment cannot be used in combination. The scope of the present invention is defined by the attached claims and their equivalents. Without departing from the scope of the present invention, those skilled in the art may make various substitutions and modifications, which should all fall within the scope of the present invention.

Claims

1. A method for determining medical information, characterized in that: include: Performing image acquisition on the acquired medical page information to obtain a medical page image, wherein the medical page information records medical data related to the user; Identify text data in the medical page image to obtain initial page text data; Performing word segmentation processing on the initial page text data based on a preset word segmentation rule to obtain page text data, wherein the page text data includes initial medical attribute description text and initial medical attribute data corresponding to the initial medical attribute description text, and the initial medical attribute description text includes at least one of physical sign indicator description data, physical examination indicator description data, and symptom indicator description data; Inputting the page text data into a medical data update model constructed based on a large language model, and outputting target text data, wherein the target text data includes a target medical attribute name and target medical attribute data corresponding to the initial medical attribute description text, the target medical attribute data has a standardized target data attribute corresponding to the target medical attribute name, and the target data attribute includes at least one of a data value threshold, a data value unit identifier, and a data type; and Target medical information representing the medical condition of the user is generated according to the target medical attribute name and the target medical attribute data.

2. The method according to claim 1, characterized in that The medical data update model is trained based on the following training method, which includes: Acquire a training sample, wherein the training sample includes sample medical attribute description data and sample medical attribute data, as well as a label medical attribute name and label medical attribute data; Input the sample medical attribute description data and the sample medical attribute data into the initial medical data update model, and output the sample target medical attribute name and the sample target medical attribute data; The initial medical data update model is trained according to the sample target medical attribute name and the sample target medical attribute data, and the label medical attribute name and the label medical attribute data to obtain the medical data update model.

3. The method according to claim 2, characterized in that The obtaining of training samples comprises: Performing text recognition on the acquired sample medical page image to obtain the sample medical attribute description data and sample medical attribute data; Determining the label medical attribute name associated with the sample medical attribute description data from a preset standard medical attribute data table; and The sample medical attribute data is updated according to the indicator data attribute associated with the label medical attribute name to obtain the label medical attribute data.

4. The method according to claim 2, characterized in that: The training method also includes: Perform a desensitization operation on the training sample to obtain a desensitized training sample.

5. The method according to any one of claims 1 to 4, characterized in that Also includes: Send a login request to the medical system; as well as Upon receiving a successful login message sent by the medical system, the medical page information is obtained from the medical system.

6. A medical information determination device, characterized in that: include: A text recognition module, used for performing image acquisition on the acquired medical page information to obtain a medical page image, wherein the medical page information records medical data related to the user; Identify text data in the medical page image to obtain initial page text data; Performing word segmentation processing on the initial page text data based on a preset word segmentation rule to obtain page text data, wherein the page text data includes initial medical attribute description text and initial medical attribute data corresponding to the initial medical attribute description text, and the initial medical attribute description text includes at least one of physical sign indicator description data, physical examination indicator description data, and symptom indicator description data; a target text data acquisition module, configured to input the page text data into a medical data update model constructed based on a large language model, and output target text data, wherein the target text data includes a target medical attribute name and target medical attribute data corresponding to the initial medical attribute description text, the target medical attribute data has a standardized target data attribute corresponding to the target medical attribute name, and the target data attribute includes at least one of a data value threshold, a data value unit identifier, and a data type; and The target medical information acquisition module is used to generate target medical information representing the medical condition of the user according to the target medical attribute name and the target medical attribute data.

7. An electronic device, characterized in that: include: one or more processors; a storage device for storing one or more programs, When the one or more programs are executed by the one or more processors, the one or more processors execute the method according to any one of claims 1 to 5.

8. A computer-readable storage medium, characterized in that: Executable instructions are stored thereon, and when the instructions are executed by a processor, the processor executes the method according to any one of claims 1 to 5.

Images