CN114220542A - Physical examination information management method and device, storage medium and computing equipment - Google Patents
Physical examination information management method and device, storage medium and computing equipment Download PDFInfo
- Publication number
- CN114220542A CN114220542A CN202111531848.3A CN202111531848A CN114220542A CN 114220542 A CN114220542 A CN 114220542A CN 202111531848 A CN202111531848 A CN 202111531848A CN 114220542 A CN114220542 A CN 114220542A
- Authority
- CN
- China
- Prior art keywords
- physical examination
- report
- health
- index
- indexes
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H50/00—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics
- G16H50/30—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for calculating health indices; for individual health risk assessment
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/36—Creation of semantic tools, e.g. ontology or thesauri
- G06F16/367—Ontology
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H15/00—ICT specially adapted for medical reports, e.g. generation or transmission thereof
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Medical Informatics (AREA)
- Physics & Mathematics (AREA)
- Health & Medical Sciences (AREA)
- Public Health (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Software Systems (AREA)
- Evolutionary Computation (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Primary Health Care (AREA)
- Artificial Intelligence (AREA)
- General Health & Medical Sciences (AREA)
- Databases & Information Systems (AREA)
- Life Sciences & Earth Sciences (AREA)
- Epidemiology (AREA)
- Evolutionary Biology (AREA)
- Pathology (AREA)
- Biomedical Technology (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Animal Behavior & Ethology (AREA)
- Computational Linguistics (AREA)
- Medical Treatment And Welfare Office Work (AREA)
Abstract
本发明提供了一种体检信息管理方法、装置、存储介质及计算设备，该方法包括：获取目标对象的至少一体检报告；所述体检报告包括电子版体检报告和/或纸质版体检报告；对所述体检报告进行解析，以得到每份所述体检报告包含的体检指标；将所述体检指标输入已训练的健康风险预测模型，利用所述健康风险预测模型对所述体检指标进行学习，以输出所述目标对象的健康风险信息；在医学知识图谱中搜索与所述健康风险信息匹配的健康管理方案。本发明提供的方案不仅可以辅助用户了解自身的健康状况，还可以向目标用户提供相应健康管理方案，辅助用户进行自身健康对管理，为用户提供极大的便利性。
The present invention provides a physical examination information management method, device, storage medium and computing device. The method includes: obtaining at least one physical examination report of a target object; the physical examination report includes an electronic version of the physical examination report and/or a paper version of the physical examination report; Analyzing the physical examination report to obtain the physical examination indicators included in each of the physical examination reports; input the physical examination indicators into a trained health risk prediction model, and use the health risk prediction model to learn the physical examination indicators, to output the health risk information of the target object; search the medical knowledge graph for a health management solution that matches the health risk information. The solution provided by the present invention can not only assist users to understand their own health status, but also provide corresponding health management solutions to target users, assist users to manage their own health, and provide users with great convenience.
Description
Technical Field
The invention relates to the technical field of artificial intelligence and intelligent medical treatment, in particular to a physical examination information management method, a physical examination information management device, a physical examination information storage medium and computing equipment.
Background
With the continuous improvement of living standard and cultural quality of people, people pay more and more attention to the living quality, and whether the health of people directly influences the living quality of people. Regular health examinations are an effective way to understand the health of a body. Physical examination can be carried out, the physical condition of the user can be mastered at any time, the health file of the user is established, and the user can find and take measures in time in advance; can be used for preventing and treating diseases in early stage, effectively reduces the expense of medical expenses, better improves the life quality and the work efficiency of people and keeps people in a healthy state.
In most cases, after each physical examination, a user generates an independent physical examination report, different physical examination institutions have different corresponding physical examination items, and the same physical examination index may correspond to different medical terms, which causes great inconvenience for the client to know the health condition of the client and also fails to provide a reasonable health management suggestion for the user.
Disclosure of Invention
In view of the above problems, the present invention has been made to provide a physical examination information management method, apparatus, storage medium, and computing device that overcome the above problems or at least partially solve the above problems.
According to a first aspect of the present invention, there is provided a physical examination information management method, the method including:
acquiring at least one physical examination report of a target object; the physical examination report comprises an electronic plate body examination report and/or a paper plate body examination report;
analyzing the physical examination reports to obtain physical examination indexes contained in each physical examination report;
inputting the physical examination indexes into a trained health risk prediction model, and learning the physical examination indexes by using the health risk prediction model to output health risk information of the target object;
searching a medical knowledge graph for a health management program matching the health risk information.
Optionally, the parsing the physical examination report to obtain the physical examination indexes included in each physical examination report includes:
identifying textual information in the physical examination report;
performing text analysis on the text information to acquire physical examination indexes contained in the text information;
and standardizing the index names of the physical examination indexes by using a medical knowledge map, and integrating and merging the physical examination indexes subjected to the standardized index names.
Optionally, the identifying text information in the physical examination report includes:
screening first text information contained in the physical examination report as the text information; and/or the presence of a gas in the gas,
and performing character detection on the image information in the physical examination report by using an Optical Character Recognition (OCR) technology to identify second text information contained in the image information as the text information.
Optionally, the method further comprises:
acquiring medical knowledge data from a plurality of medical knowledge data sources;
extracting a plurality of entities in the medical knowledge data, and learning an incidence relation between the entities;
constructing a medical knowledge graph by using the plurality of entities and the incidence relation between the entities; in the medical knowledge graph, one entity corresponds to one entity node, each entity node is provided with a plurality of characteristic attributes, and the characteristic attributes comprise medical standard names corresponding to the entities and medical aliases of the medical standard names.
Optionally, after the constructing the medical knowledge graph by using the plurality of entities and the association relationship between the entities, the method further includes:
creating a plurality of risk factor nodes corresponding to different disease risk factors in the medical knowledge graph;
and acquiring a health management scheme corresponding to each disease risk factor, and taking the health management scheme as a characteristic attribute associated with the risk factor node.
Optionally, the normalizing the index name of the physical examination index by using the medical knowledge graph, and the integrating and merging the physical examination index subjected to the normalizing process of the index name include:
acquiring a plurality of physical examination indexes contained in each physical examination report, determining an entity node corresponding to each physical examination index by using the characteristic attribute of each entity node in the medical knowledge graph for each physical examination index, and acquiring a medical standard name corresponding to each entity node as an index name corresponding to each physical examination index;
and determining at least one physical examination index which corresponds to the same entity node and has the same index name in each physical examination report, and integrating and merging the at least one physical examination index which corresponds to the same entity node.
Optionally, before the inputting the physical examination indicator into the trained health risk prediction model, the method further comprises:
constructing a logistic regression model;
collecting health related data corresponding to a plurality of reference objects, and generating a sample data set for model training by using the health related data; the sample data set comprises a plurality of groups of data pairs, and each group of data pairs comprises body index data matched with physical examination indexes and corresponding disease types;
dividing the sample data set into a training set and a test set;
performing model training on a logistic regression model by using the training set to obtain model parameters corresponding to the logistic regression model;
generating a health risk prediction model for health risk prediction by combining the model parameters and the logistic regression model;
testing and optimizing the health risk prediction model using the test set.
According to a second aspect of the present invention, there is also provided a physical examination information management apparatus, the apparatus including:
the report acquisition module is used for acquiring at least one physical examination report of the target object; the physical examination report comprises an electronic plate body examination report and/or a paper plate body examination report;
the report analysis module is used for analyzing the physical examination reports to obtain physical examination indexes contained in each physical examination report;
the model learning module is used for inputting the physical examination indexes into a trained health risk prediction model, and learning the physical examination indexes by using the health risk prediction model so as to output health risk information of the target object;
and the scheme matching module is used for searching the health management scheme matched with the health risk information in the medical knowledge graph.
According to a third aspect of the present invention, there is provided a computer readable storage medium for storing program code for performing the method of any one of the first aspect.
According to a fourth aspect of the present invention, there is also provided a computing device comprising a processor and a memory:
the memory is used for storing program codes and transmitting the program codes to the processor;
the processor is configured to perform the method of any of the first aspect according to instructions in the program code.
The invention provides a physical examination information management method, a physical examination information management device, a storage medium and computing equipment. In addition, the physical examination indexes of the target object can be learned, corresponding disease risk types and probabilities can be given according to the past physical examination indexes, the user can be warned of the health in time, finally, a corresponding health management scheme can be provided for the target user, the user can be assisted in carrying out self health management, and great convenience is provided for the user.
The foregoing description is only an overview of the technical solutions of the present invention, and the embodiments of the present invention are described below in order to make the technical means of the present invention more clearly understood and to make the above and other objects, features, and advantages of the present invention more clearly understandable.
The above and other objects, advantages and features of the present invention will become more apparent to those skilled in the art from the following detailed description of specific embodiments thereof, taken in conjunction with the accompanying drawings.
Drawings
Various other advantages and benefits will become apparent to those of ordinary skill in the art upon reading the following detailed description of the preferred embodiments. The drawings are only for purposes of illustrating the preferred embodiments and are not to be construed as limiting the invention. Also, like reference numerals are used to refer to like parts throughout the drawings. In the drawings:
fig. 1 shows a flow diagram of a physical examination information management method according to an embodiment of the present invention;
FIG. 2 is a schematic diagram illustrating a process for text recognition in a physical examination report according to an embodiment of the present invention;
fig. 3 is a schematic structural view showing a physical examination information management apparatus according to an embodiment of the present invention;
fig. 4 is a schematic view showing the construction of a physical examination information management apparatus according to another embodiment of the present invention;
FIG. 5 shows a schematic diagram of a computing device architecture according to an embodiment of the invention.
Detailed Description
Exemplary embodiments of the present invention will be described in more detail below with reference to the accompanying drawings. While exemplary embodiments of the invention are shown in the drawings, it should be understood that the invention can be embodied in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the invention to those skilled in the art.
The solution provided in this embodiment mainly combines with an Artificial Intelligence (AI) technology to acquire and process related data, wherein the AI is a theory, method, technique, and application system that simulates, extends, and expands human Intelligence using a digital computer or a machine of a digital computer controller, senses the environment, acquires knowledge, and uses the knowledge to obtain the best result.
The artificial intelligence technology base generally comprises technologies such as sensors, special artificial intelligence chips, cloud computing, distributed storage, big data processing technology, operation/interaction systems, electromechanical integration and the like. The artificial intelligence software technology mainly comprises a computer vision technology, a robot technology, a biological recognition technology, a voice processing technology, a natural voice processing technology, machine learning/deep learning and the like.
As can be seen from fig. 1, the method provided by the embodiment of the present invention at least includes the following steps S101 to S104.
S101, acquiring at least one integrated inspection report of a target object; the physical examination report comprises an electronic plate physical examination report and/or a paper plate physical examination report.
The target object in this embodiment may be any user object that has completed physical examination, and the physical examination report of the target object may be a paper-based physical examination report or an electronic-based physical examination report. When the physical examination report is an electronic physical examination report, the physical examination report can be acquired from a third-party physical examination institution which performs physical examination on the target object, or the target object can be uploaded to acquire the physical examination report. Optionally, the physical examination report can be acquired through an electronic physical examination report hyperlink input by the target object. When the physical examination report is a paper-based physical examination report, image information corresponding to the paper-based physical examination report can be acquired by using image acquisition equipment (such as a camera, a camera and the like) to acquire the physical examination report.
In practical applications, the physical examination reports corresponding to the target object may be multiple, when there are multiple physical examination reports of the target object, the multiple physical examination reports of the target object may be obtained from different channels, and the multiple physical examination reports may be within a certain time range, such as within two years, within five years, and the like, or all the physical examination reports during the historical physical examination of the target object, which is not limited in the embodiment of the present invention.
S102, analyzing the physical examination reports to obtain physical examination indexes contained in each physical examination report.
Each acquired physical examination report can be analyzed to obtain the physical examination indexes contained in the physical examination report.
In an optional embodiment of the present invention, when the step S102 acquires the physical examination index in the physical examination report, the method may further include:
s102-1, identifying text information in the physical examination report. As described above, the physical examination report may include a paper-based physical examination report or an electronic-based physical examination report, and different methods are used for acquiring text information in the physical examination reports for different types of physical examination reports.
Optionally, for the electronic plate physical examination report, the first text information included in the physical examination report may be screened as the text information. That is, for the electronic physical examination report, the text information can be directly screened out.
For a paper-based physical examination report, the image information in the physical examination report may be subjected to text detection by using an Optical Character Recognition (OCR) technology to identify second text information included in the image information as the text information. For paper-based physical examination reports, since they were last time in the form of pictures, it is necessary to recognize text information included in an image using an optical character recognition OCR technology.
Optical Character Recognition (OCR) refers to a process of analyzing and recognizing an image file of text data to obtain text and layout information. I.e. the text in the image is recognized and returned in the form of text. A typical OCR technology route is shown in figure 2 below.
The image pre-processing stage uses a CNN-based neural network as a feature extraction means. The strong learning ability of the CNN is benefited, the robustness of feature extraction can be enhanced by matching with a large amount of data, and good robustness can be shown when the image problems of blurring, distortion, complex background, unclear light and the like are encountered.
Character detection is to detect the position and range of the text and its layout. Layout analysis and text line detection are also typically included. The main problem to be solved by character detection is where the characters are, and how large the range of the characters is.
The text recognition is to recognize the text content on the basis of text detection and convert the text information in the image into text information. The main problem to be solved by word recognition is what each word is. The recognized text typically needs to be checked again to ensure its correctness.
In practical applications, the electronic version physical examination report may also include an image, and the two methods may be simultaneously adopted for the electronic version physical examination report to identify text information included in the physical examination report, and may be specifically set according to different requirements, which is not limited in the embodiment of the present invention.
S102-2, performing text analysis on the text information to obtain physical examination indexes contained in the text information.
The collected text information can be further analyzed to obtain physical examination indexes contained in the text information. The physical examination indexes contained in the text information can be multiple items, and during specific identification, the physical examination indexes in the physical examination report can be identified by combining a keyword detection technology, a text identification technology and the like. The physical examination index may include an index name and a corresponding detection value.
S102-3, standardizing the index names of the physical examination indexes by using a medical knowledge map, and integrating and merging the physical examination indexes subjected to the standardized index names.
For the physical examination reports issued by different organizations, the technical terms used by the physical examination reports may be different, and therefore, in this embodiment, the medical knowledge map may be further used to perform the standardization processing of the index name on the physical examination indexes, so as to integrate and merge the physical examination indexes after the standardization processing of the index name.
Before this, the medical knowledge map is created, and the specific process is as follows, steps A1-A3.
A1, acquiring medical knowledge data from a plurality of medical knowledge data sources. In particular, data may be crawled from various sources of empowerment medical knowledge data.
A2, extracting a plurality of entities in the medical knowledge data, and learning the association relation between the entities. And extracting entities and relations from the data acquired in the step A1, wherein the entities and relations comprise various medical term names, such as hemoglobin, platelets and other entities. The entity relationship may be whether each entity has an association relationship or an interaction relationship.
A3, constructing a medical knowledge map by using the plurality of entities and the incidence relation between the entities; in the medical knowledge graph, one entity corresponds to one entity node, each entity node is provided with a plurality of characteristic attributes, and the characteristic attributes comprise medical standard names corresponding to the entities and medical aliases of the medical standard names.
After the entities and the entity relationships are obtained, the entities are used as nodes, and the entity relationships can be used as connecting lines to establish a medical knowledge graph. In an optional embodiment of the invention, the medical knowledge graph can be released to a designated platform, feedback information of medical experts aiming at the medical knowledge graph is collected, and then the iterative medical knowledge graph is continuously optimized by using the feedback information until the result reaches the set precision. In the medical knowledge graph constructed in this embodiment, the physical examination index is a node, each node has a plurality of characteristic attributes, and the medical standard name corresponding to the entity node and the medical alias of the medical standard name are used as one of the attributes.
After the medical knowledge graph is established, the medical knowledge graph is used for carrying out standardization processing on the physical examination indexes, that is, a plurality of physical examination indexes contained in each physical examination report are obtained, for each physical examination index, an entity node corresponding to the physical examination index is determined by using the characteristic attribute of each entity node in the medical knowledge graph, and a medical standard name corresponding to the entity node is obtained to be used as an index name corresponding to the physical examination index; and determining at least one physical examination index which corresponds to the same entity node and has the same index name in each physical examination report, and integrating and merging the at least one physical examination index which corresponds to the same entity node.
For example, glycated hemoglobin may be referred to as HbA1c, HGB, etc., hemoglobin may be referred to as hemoglobin, and reports from different health institutions may be referred to by different names. After the index fields on the physical examination reports are identified by using an OCR technology, matching is carried out in the knowledge graph, entity nodes corresponding to the physical examination indexes in the knowledge graph are found, and the entity nodes are converted into standard names of the physical examination indexes. For example, assume that a physical examination index 1-HbA1 c-detection value a is described in physical examination report 1; the physical examination report 2 describes a physical examination index 2-HGB-detection value b; the physical examination report 3 describes the physical examination index 3, namely, the glycated hemoglobin-detected value c, and when the physical examination reports 1, 2, and 3 are integrated, the physical examination indexes 1 to HbA1c, 2 to HGB, and 3 in the physical examination reports are glycated hemoglobin, and in this case, the physical examination indexes 1 to HbA1c, 2 to HGB, and 3 may be integrated, and the physical examination indexes 1 to HbA1c and 2 to HGB may be modified to glycated hemoglobin.
When at least one physical examination index corresponding to the same entity node is integrated and merged, detection values corresponding to the unified physical examination index in physical examination reports issued by different physical examination mechanisms at different times can be summarized in a list or other manners, for example, the form after the physical examination index a is summarized can include: at time point 1, the physical examination mechanism 1 detects a value 1; time point 2, physical examination institution 2, detection 2.; at time 2, physical examination facility n, detects value n. Specifically, a plurality of physical examination indexes corresponding to the same physical node can be sorted according to the time sequence of the physical examination reports, and then the integration and merging are completed.
S103, inputting the physical examination indexes into a trained health risk prediction model, and learning the physical examination indexes by using the health risk prediction model to output health risk information of the target object.
After the physical examination reports are obtained and subjected to physical examination index identification and integration, the physical examination reports after integration can be input into a health risk prediction model and trained for learning. The health risk prediction model of this embodiment is an intelligent learning model, input data of the model is various physical examination indexes, output data is health risk information matched with the input physical examination indexes, and the health risk information may include a risk type and a corresponding risk probability.
Before the physical examination indexes are learned by using the health risk prediction model, the health risk prediction model can be constructed and trained in advance, and then the trained health risk prediction model is used for intelligent learning. Optionally, the training health risk prediction model provided in this embodiment may specifically include the following steps B1-B
And B1, constructing a logistic regression model.
The logistic regression model can be considered as a linear regression model normalized by Sigmoid function (logistic equation). The most basic way to solve the parameters of logistic regression is the gradient descent method, which only needs to write the cost function and the gradient formula of the parameter θ. In this embodiment, a logistic regression model is constructed as a basic model corresponding to the health risk prediction model.
B2, collecting health related data corresponding to a plurality of reference objects, and generating a sample data set for model training by using the health related data; the sample data set comprises a plurality of groups of data pairs, and each group of data pairs comprises body index data matched with physical examination indexes and corresponding disease types.
The reference object in this embodiment may be any user or patient, and the health related data corresponding to the reference object may include health index data, historical disease parameters, and real data related to the physical health of the reference object in multiple time periods. For the obtained health related data, a sample data set for model training can be constructed by using the obtained health related data. Optionally, the collected health-related data may be subjected to data cleansing to extract key information in the health-related data. Specifically, the entity nodes in the medical knowledge graph can be used as the screening keywords, and for any reference object, specific data associated with each physical examination index is selected from health related data corresponding to the reference object. The method provided by the embodiment can be used for cleaning and preprocessing the data by combining the medical knowledge graph, so that the sample data can be standardized to correct missing values, misspelling and normalization/standardization of numerical values to make the data have the problems of comparability, data conversion (such as logarithmic conversion) and the like, and the training is more accurate to the model.
For any reference object, in addition to the corresponding physical examination index and the data related thereto, it is also necessary to obtain the health condition information corresponding to the reference object, for example, if there is a disease, it is also necessary to determine the type of the disease and the related information such as the risk level.
B3, dividing the sample data set into a training set and a test set.
For the collected sample data set, a training set and a test set can be divided. Specifically, during the division, 70% of data in the sample data set can be selected in a random manner to construct a training set, and the remaining 30% of data can be used to construct a test set.
That is, a prediction model is built using a training set, and then such a trained model is applied to a test set (i.e., as new, unseen data) for prediction. The best model is selected according to the performance of the model on the test set, and hyper-parameter optimization can be carried out in order to obtain the best model. In addition, the sample data set may be divided into a training set, a validation set, and a test set. Similar to the above explanation, the training set is used to build a prediction model, and the validation set is evaluated, so that prediction can be performed, model tuning (e.g., hyper-parametric optimization) can be performed, and the model with the best performance can be selected according to the result of the validation set.
And B4, performing model training on the logistic regression model by using the training set to obtain model parameters corresponding to the logistic regression model.
Specifically during the training process, the dataset can be segmented into N folds using N-fold Cross Validation (CV) (i.e., typically using 5-fold or 10-fold CV). In such an N-fold CV, one of the folds is retained as test data, while the remaining folds are used as training data for modeling.
B5, generating a health risk prediction model for health risk prediction by combining the model parameters and the logistic regression model;
b6, testing and optimizing the health risk prediction model using the test set.
The hyper-parameters are essentially parameters of a machine learning algorithm, and directly influence the learning process and the prediction performance, so that the hyper-parameters in the model need to be optimized, and the optimized health risk prediction model is obtained.
In the embodiment of the invention, in the process of machine learning of the health risk prediction model, besides learning the disease types corresponding to different physical examination indexes, the change rule of the physical examination indexes in a certain time period can be learned, and further the relation between the change rule of the physical examination indexes and the disease types can be learned.
That is, when constructing the sample data set, for each group of sample data, a time dimension may be added, and a change rule of the physical examination index within a certain time is recorded, for example, a change value of the physical examination index within a continuous time, such as gradually increasing, gradually decreasing, and the like, so as to find an association relationship between the change rule of the physical examination index and the disease.
After the physical examination indexes corresponding to the target object are input into the health risk prediction model, the corresponding disease types and the disease probabilities can be predicted according to the abnormity of the physical examination indexes, and meanwhile, the change rules of the physical examination indexes can be used as prediction bases, so that the prediction results of the disease risk prediction model are more accurate.
And S104, searching a health management scheme matched with the health risk information in the medical knowledge graph.
After determining the health risk information corresponding to the target object by using the disease risk prediction model, the health management scheme matched with the health risk information can be searched in the medical knowledge map.
In the foregoing, in the medical knowledge graph, entities corresponding to different physical examination indicators may be used as nodes, and further, nodes corresponding to disease risk factors may be created in the medical knowledge graph. Specifically, a plurality of risk factor nodes corresponding to different disease risk factors may be created in the medical knowledge graph; and acquiring a health management scheme corresponding to each disease risk factor, and taking the health management scheme as a characteristic attribute associated with the risk factor node.
In other words, in the medical knowledge-graph, the disease risk factor is used as a node, and the corresponding management scheme is an attribute (such as a sugar control scheme) of the node. Each disease also exists as a node in the medical knowledge map, and its corresponding health management plan is also an attribute of the disease, (e.g., diabetes management plan is an attribute of diabetes). If the health risk prediction model predicts that the diabetes risk of the target object is high and the fasting blood glucose is one of the risk factors, the medical knowledge graph is searched for the fasting blood glucose and the diabetes, and then the corresponding management scheme can be found. Of course, the feature attributes corresponding to the risk factor node corresponding to any disease risk factor may include features such as influence factors of the disease risk factor, in addition to the associated health management scheme, which is not limited in this embodiment.
The embodiment of the invention provides a physical examination information management method, which integrates different physical examination reports of the same object, integrates corresponding same physical examination indexes in the different physical examination reports, and can assist a user in knowing the self health condition. In addition, the physical examination indexes of the target object can be learned, corresponding disease risk types and probabilities can be given according to the past physical examination indexes, the user can be warned of the health in time, finally, a corresponding health management scheme can be provided for the target user, the user can be assisted in carrying out self health management, and great convenience is provided for the user.
Based on the unified inventive concept, an embodiment of the present invention further provides a physical examination information management apparatus, as shown in fig. 3, the physical examination information management apparatus provided in this embodiment may include:
a report acquisition module 310, configured to acquire at least one physical examination report of a target subject; the physical examination report comprises an electronic plate body examination report and/or a paper plate body examination report;
a report parsing module 320, configured to parse the physical examination reports to obtain physical examination indicators included in each physical examination report;
a model learning module 330, configured to input the physical examination indicator into a trained health risk prediction model, and learn the physical examination indicator by using the health risk prediction model to output health risk information of the target object;
and the scheme matching module 340 is used for searching the medical knowledge graph for the health management scheme matched with the health risk information.
In an optional embodiment of the present invention, the report parsing module 320 may further be configured to:
identifying textual information in the physical examination report;
performing text analysis on the text information to acquire physical examination indexes contained in the text information;
and standardizing the index names of the physical examination indexes by using a medical knowledge map, and integrating and merging the physical examination indexes subjected to the standardized index names.
In an optional embodiment of the present invention, the report parsing module 320 may further be configured to:
screening first text information contained in the physical examination report as the text information; and/or the presence of a gas in the gas,
and performing character detection on the image information in the physical examination report by using an Optical Character Recognition (OCR) technology to identify second text information contained in the image information as the text information.
In an alternative embodiment of the present invention, as shown in fig. 4, the physical examination information management apparatus may further include a knowledge graph creation module 350;
a knowledge-graph creation module 350 for collecting medical knowledge data from a plurality of medical knowledge data sources;
extracting a plurality of entities in the medical knowledge data, and learning an incidence relation between the entities;
constructing a medical knowledge graph by using the plurality of entities and the incidence relation between the entities; in the medical knowledge graph, one entity corresponds to one entity node, each entity node is provided with a plurality of characteristic attributes, and the characteristic attributes comprise medical standard names corresponding to the entities and medical aliases of the medical standard names.
In an optional embodiment of the present invention, the knowledge-graph creation module 350 may be further configured to:
creating a plurality of risk factor nodes corresponding to different disease risk factors in the medical knowledge graph;
and acquiring a health management scheme corresponding to each disease risk factor, and taking the health management scheme as a characteristic attribute associated with the risk factor node.
In an optional embodiment of the present invention, the report parsing module 320 may further be configured to:
acquiring a plurality of physical examination indexes contained in each physical examination report, determining an entity node corresponding to each physical examination index by using the characteristic attribute of each entity node in the medical knowledge graph for each physical examination index, and acquiring a medical standard name corresponding to each entity node as an index name corresponding to each physical examination index;
and determining at least one physical examination index which corresponds to the same entity node and has the same index name in each physical examination report, and integrating and merging the at least one physical examination index which corresponds to the same entity node.
In an alternative embodiment of the present invention, as shown in fig. 4, the physical examination information management apparatus may further include a model creation module 360;
the model creation module 360 is used for constructing a logistic regression model;
collecting health related data corresponding to a plurality of reference objects, and generating a sample data set for model training by using the health related data; the sample data set comprises a plurality of groups of data pairs, and each group of data pairs comprises body index data matched with physical examination indexes and corresponding disease types;
dividing the sample data set into a training set and a test set;
performing model training on a logistic regression model by using the training set to obtain model parameters corresponding to the logistic regression model;
generating a health risk prediction model for health risk prediction by combining the model parameters and the logistic regression model;
testing and optimizing the health risk prediction model using the test set.
An embodiment of the present invention further provides a computer-readable storage medium, where the computer-readable storage medium is used for storing a program code, and the program code is used for executing the method described in the above embodiment.
An embodiment of the present invention further provides a computing device, where the computing device includes a processor and a memory: the memory is used for storing program codes and transmitting the program codes to the processor; the processor is configured to execute the method according to the above embodiment according to the instructions in the program code. Optionally, referring to fig. 5, the computer device may further include a user interface, a network interface, a camera, Radio Frequency (RF) circuitry, sensors, audio circuitry, a WI-FI module, and so forth. The user interface may include a Display screen (Display), an input unit such as a keypad (Keyboard), etc., and the optional user interface may also include a USB interface, a card reader interface, etc. The network interface may optionally include a standard wired interface, a wireless interface (e.g., a bluetooth interface, WI-FI interface), etc.
It is clear to those skilled in the art that the specific working processes of the above-described systems, devices, modules and units may refer to the corresponding processes in the foregoing method embodiments, and for the sake of brevity, further description is omitted here.
In addition, the functional units in the embodiments of the present invention may be physically independent of each other, two or more functional units may be integrated together, or all the functional units may be integrated in one processing unit. The integrated functional units may be implemented in the form of hardware, or in the form of software or firmware.
Those of ordinary skill in the art will understand that: the integrated functional units, if implemented in software and sold or used as a stand-alone product, may be stored in a computer readable storage medium. Based on such understanding, the technical solution of the present invention may be embodied in the form of a software product, which is stored in a storage medium and includes instructions for causing a computing device (e.g., a personal computer, a server, or a network device) to execute all or part of the steps of the method according to the embodiments of the present invention when the instructions are executed. And the aforementioned storage medium includes: u disk, removable hard disk, Read Only Memory (ROM), Random Access Memory (RAM), magnetic or optical disk, and other various media capable of storing program code.
Alternatively, all or part of the steps of implementing the foregoing method embodiments may be implemented by hardware (such as a computing device, e.g., a personal computer, a server, or a network device) associated with program instructions, which may be stored in a computer-readable storage medium, and when the program instructions are executed by a processor of the computing device, the computing device executes all or part of the steps of the method according to the embodiments of the present invention.
Finally, it should be noted that: the above embodiments are only used to illustrate the technical solution of the present invention, and not to limit the same; while the invention has been described in detail and with reference to the foregoing embodiments, it will be understood by those skilled in the art that: the technical solutions described in the foregoing embodiments can be modified or some or all of the technical features can be equivalently replaced within the spirit and principle of the present invention; such modifications or substitutions do not depart from the scope of the present invention.
Claims (10)
1. A physical examination information management method, characterized by comprising:
acquiring at least one physical examination report of a target object; the physical examination report comprises an electronic plate body examination report and/or a paper plate body examination report;
analyzing the physical examination reports to obtain physical examination indexes contained in each physical examination report;
inputting the physical examination indexes into a trained health risk prediction model, and learning the physical examination indexes by using the health risk prediction model to output health risk information of the target object;
searching a medical knowledge graph for a health management program matching the health risk information.
2. The method of claim 1, wherein the parsing the physical examination reports to obtain the physical examination indexes contained in each physical examination report comprises:
identifying textual information in the physical examination report;
performing text analysis on the text information to acquire physical examination indexes contained in the text information;
and standardizing the index names of the physical examination indexes by using a medical knowledge map, and integrating and merging the physical examination indexes subjected to the standardized index names.
3. The method of claim 2, wherein the identifying textual information in the physical examination report comprises:
screening first text information contained in the physical examination report as the text information; and/or the presence of a gas in the gas,
and performing character detection on the image information in the physical examination report by using an Optical Character Recognition (OCR) technology to identify second text information contained in the image information as the text information.
4. The method of claim 2, further comprising:
acquiring medical knowledge data from a plurality of medical knowledge data sources;
extracting a plurality of entities in the medical knowledge data, and learning an incidence relation between the entities;
constructing a medical knowledge graph by using the plurality of entities and the incidence relation between the entities; in the medical knowledge graph, one entity corresponds to one entity node, each entity node is provided with a plurality of characteristic attributes, and the characteristic attributes comprise medical standard names corresponding to the entities and medical aliases of the medical standard names.
5. The method of claim 4, wherein the constructing the medical knowledge map using the plurality of entities and the associations between the entities further comprises:
creating a plurality of risk factor nodes corresponding to different disease risk factors in the medical knowledge graph;
and acquiring a health management scheme corresponding to each disease risk factor, and taking the health management scheme as a characteristic attribute associated with the risk factor node.
6. The method of claim 4, wherein the normalizing the physical examination indexes by index names by using the medical knowledge map and the integrating and merging the physical examination indexes after the normalizing of the index names comprises:
acquiring a plurality of physical examination indexes contained in each physical examination report, determining an entity node corresponding to each physical examination index by using the characteristic attribute of each entity node in the medical knowledge graph for each physical examination index, and acquiring a medical standard name corresponding to each entity node as an index name corresponding to each physical examination index;
and determining at least one physical examination index which corresponds to the same entity node and has the same index name in each physical examination report, and integrating and merging the at least one physical examination index which corresponds to the same entity node.
7. The method of any of claims 1-6, wherein prior to entering the physical indicators into the trained health risk prediction model, the method further comprises:
constructing a logistic regression model;
collecting health related data corresponding to a plurality of reference objects, and generating a sample data set for model training by using the health related data; the sample data set comprises a plurality of groups of data pairs, and each group of data pairs comprises body index data matched with physical examination indexes and corresponding disease types;
dividing the sample data set into a training set and a test set;
performing model training on a logistic regression model by using the training set to obtain model parameters corresponding to the logistic regression model;
generating a health risk prediction model for health risk prediction by combining the model parameters and the logistic regression model;
testing and optimizing the health risk prediction model using the test set.
8. A physical examination information management apparatus, characterized in that the apparatus comprises:
the report acquisition module is used for acquiring at least one physical examination report of the target object; the physical examination report comprises an electronic plate body examination report and/or a paper plate body examination report;
the report analysis module is used for analyzing the physical examination reports to obtain physical examination indexes contained in each physical examination report;
the model learning module is used for inputting the physical examination indexes into a trained health risk prediction model, and learning the physical examination indexes by using the health risk prediction model so as to output health risk information of the target object;
and the scheme matching module is used for searching the health management scheme matched with the health risk information in the medical knowledge graph.
9. A computer-readable storage medium, characterized in that the computer-readable storage medium is configured to store a program code for performing the method of any of claims 1-7.
10. A computing device, the computing device comprising a processor and a memory:
the memory is used for storing program codes and transmitting the program codes to the processor;
the processor is configured to perform the method of any of claims 1-7 according to instructions in the program code.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111531848.3A CN114220542A (en) | 2021-12-15 | 2021-12-15 | Physical examination information management method and device, storage medium and computing equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111531848.3A CN114220542A (en) | 2021-12-15 | 2021-12-15 | Physical examination information management method and device, storage medium and computing equipment |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114220542A true CN114220542A (en) | 2022-03-22 |
Family
ID=80702159
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111531848.3A Pending CN114220542A (en) | 2021-12-15 | 2021-12-15 | Physical examination information management method and device, storage medium and computing equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114220542A (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116150426A (en) * | 2023-02-21 | 2023-05-23 | 上海哔哩哔哩科技有限公司 | Video manuscript-based processing method and device, storage medium and electronic equipment |
| CN116246750A (en) * | 2023-03-06 | 2023-06-09 | 吾征智能技术(北京)有限公司 | Physical examination report reading system and method |
| CN117809798A (en) * | 2024-03-01 | 2024-04-02 | 金堂县第一人民医院 | Inspection report interpretation methods, systems, equipment and media based on large models |
| WO2024067442A1 (en) * | 2022-09-27 | 2024-04-04 | 华为技术有限公司 | Data management method and related apparatus |
| CN119905194A (en) * | 2025-03-27 | 2025-04-29 | 成都企康科技有限公司 | Standardized processing method, device, electronic device and storage medium for physical examination data |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106021281A (en) * | 2016-04-29 | 2016-10-12 | 京东方科技集团股份有限公司 | Method for establishing medical knowledge graph, device for same and query method for same |
| CN112037910A (en) * | 2020-08-31 | 2020-12-04 | 康键信息技术（深圳）有限公司 | Health information management method, device, equipment and storage medium |
| CN113380380A (en) * | 2021-06-23 | 2021-09-10 | 上海电子信息职业技术学院 | Intelligent reading device for medical reports |
-
2021
- 2021-12-15 CN CN202111531848.3A patent/CN114220542A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106021281A (en) * | 2016-04-29 | 2016-10-12 | 京东方科技集团股份有限公司 | Method for establishing medical knowledge graph, device for same and query method for same |
| CN112037910A (en) * | 2020-08-31 | 2020-12-04 | 康键信息技术（深圳）有限公司 | Health information management method, device, equipment and storage medium |
| CN113380380A (en) * | 2021-06-23 | 2021-09-10 | 上海电子信息职业技术学院 | Intelligent reading device for medical reports |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2024067442A1 (en) * | 2022-09-27 | 2024-04-04 | 华为技术有限公司 | Data management method and related apparatus |
| CN116150426A (en) * | 2023-02-21 | 2023-05-23 | 上海哔哩哔哩科技有限公司 | Video manuscript-based processing method and device, storage medium and electronic equipment |
| CN116246750A (en) * | 2023-03-06 | 2023-06-09 | 吾征智能技术(北京)有限公司 | Physical examination report reading system and method |
| CN117809798A (en) * | 2024-03-01 | 2024-04-02 | 金堂县第一人民医院 | Inspection report interpretation methods, systems, equipment and media based on large models |
| CN117809798B (en) * | 2024-03-01 | 2024-04-26 | 金堂县第一人民医院 | Verification report interpretation method, system, equipment and medium based on large model |
| CN119905194A (en) * | 2025-03-27 | 2025-04-29 | 成都企康科技有限公司 | Standardized processing method, device, electronic device and storage medium for physical examination data |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| JP7714617B2 (en) | Automated information extraction and refinement in pathology reports using natural language processing | |
| JP7522167B2 (en) | Systems and methods for model-assisted cohort selection | |
| CN114220542A (en) | Physical examination information management method and device, storage medium and computing equipment | |
| CN114550946B (en) | Medical data processing method, device and storage medium | |
| CN112149414B (en) | Text similarity determination method, device, equipment and storage medium | |
| CN112786194A (en) | Medical image diagnosis guide inspection system, method and equipment based on artificial intelligence | |
| Jeon et al. | Measuring the novelty of scientific publications: A fastText and local outlier factor approach | |
| US20230298589A1 (en) | Ai platform for processing speech and video information collected during a medical procedure | |
| CN112541066A (en) | Text-structured-based medical and technical report detection method and related equipment | |
| CN115841861A (en) | Similar medical record recommendation method and system | |
| CN113722507A (en) | Hospital cost prediction method and device based on knowledge graph and computer equipment | |
| CN120221058A (en) | Traditional Chinese Medicine Rehabilitation Diagnosis System Based on Multimodal Knowledge Graph and Large Language Model | |
| CN112579444A (en) | Text cognition-based automatic analysis modeling method, system, device and medium | |
| Rahman | Optimizing customer satisfaction through sentiment analysis: A BERT-based machine learning approach to extract insights | |
| CN112732910A (en) | Cross-task text emotion state assessment method, system, device and medium | |
| CN120048417A (en) | Multi-source medical clinic data comprehensive management method and platform based on RPA | |
| CN119271651A (en) | Medical knowledge base verification method, system, medium and equipment | |
| CN120579548A (en) | Medical record information extraction and analysis method and system based on deep learning large model | |
| Chondrogiannis et al. | An Intelligent Ontology Alignment Tool Dealing with Complicated Mismatches. | |
| CN114429822B (en) | Medical record quality inspection method, device and storage medium | |
| CN119153127B (en) | Infectious disease data intelligent coding method, device, computer equipment and storage medium | |
| Prashanthi et al. | Automated categorization of systemic disease and duration from electronic medical record system data using finite-state machine modeling: Prospective validation study | |
| CN116913527B (en) | Hypertension evaluation method and system based on multi-round dialogue frame | |
| Roy et al. | Bilstm models with and without pretrained embeddings and bert on german patient reviews | |
| Li et al. | DIPE: a diagnosis-assisted inquiry point extractor towards medical dialogues: Q. Li et al. |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20220322 |
|
| RJ01 | Rejection of invention patent application after publication |