A disease teaching relationship determination method and related device

Abstract

The application provides a disease teaching relationship determination method and related device, the method comprises the steps of: constructing a disease knowledge graph; determining a plurality of disease relationships corresponding to a plurality of historical case data according to the disease knowledge graph; determining a plurality of target disease relationships corresponding to a target core treatment from the plurality of disease relationships; determining a plurality of feature correlation degrees corresponding to the plurality of target disease relationships respectively; if the feature correlation degree meets a preset correlation degree threshold, the target disease relationship corresponding to the feature correlation degree is taken as a target disease teaching relationship for the target core treatment. Through the above method, the target teaching disease relationship for the target core treatment can be obtained, so that the teaching of related diseases and features for the target core treatment can be carried out.

Description

Technical Field

[0001] This application relates to the field of data, specifically to a method and related apparatus for determining disease teaching relationships. Background Technology

[0002] Clinical reasoning refers to the process from analyzing symptoms to diagnosing a disease and then proposing a treatment plan. In order to train clinical reasoning, it is necessary to use disease teaching data for teaching.

[0003] The disease teaching data in related technologies mainly includes the relationship between diseases and characteristics. This allows for training in clinical thinking, specifically the part from analyzing symptoms to diagnosing diseases. However, the disease teaching data in related technologies only includes the part from analyzing symptoms to diagnosing diseases, and does not include the part from diagnosing diseases to providing treatment plans. In other words, the disease teaching data in related technologies cannot be used to teach treatment plans. Summary of the Invention

[0004] This application provides a method and related apparatus for determining disease teaching relationships, which can obtain target teaching disease relationships for target core treatments, so as to enable teaching of related diseases and characteristics for target core treatments.

[0005] In view of this, the first aspect of the embodiments of this application provides a method for determining disease teaching relationships, including:

[0006] Construct a disease knowledge graph, which includes entities and entity relationships between entities. The entities include disease type entities, feature type entities, and treatment type entities. The treatment type entities include core treatment type entities.

[0007] Based on the disease knowledge graph, multiple disease relationships are determined corresponding to multiple historical case data. The disease relationship refers to the relationship between disease, characteristics and core treatment.

[0008] Identify multiple target disease relationships corresponding to the target core treatment from among the multiple disease relationships described above;

[0009] Determine multiple feature correlations corresponding to multiple target disease relationships, wherein the feature correlations are used to represent the correlation between the features corresponding to the target disease relationships and the target core treatment;

[0010] If the feature correlation meets the preset correlation threshold, the target disease relationship corresponding to the feature correlation is used as the target disease teaching relationship for the target core treatment.

[0011] Optionally, determining the correlation of multiple features corresponding to the multiple target disease relationships includes:

[0012] For the i-th target disease relationship among multiple target disease relationships, obtain the feature probability of the i-th target disease relationship having a corresponding feature, the target core treatment probability of taking the target core treatment, and the target core treatment conditional probability of taking the target core treatment when the i-th target disease relationship has a corresponding feature;

[0013] By inputting the feature probability, the target core treatment probability, and the target core treatment conditional probability into the following formula, the feature relevance corresponding to the i-th target disease relationship is obtained:

[0014]

[0015] Among them, P i P(A) refers to the feature relevance corresponding to the i-th target disease relationship. i P(B) refers to the feature probability of the i-th target disease relationship corresponding to the feature, and P(B|A) refers to the target core treatment probability of taking the target core treatment. i ) refers to the conditional probability of taking the target core treatment when there is an i-th feature corresponding to the target disease relationship, where i is a positive integer.

[0016] Optional methods for determining disease-based teaching relationships include:

[0017] Acquire evidence-based medicine data, which refers to data used in evidence-based medicine;

[0018] The evidence-based medicine data was segmented to obtain multiple evidence-based medicine terms;

[0019] Based on multiple evidence-based medicine terms, a disease knowledge graph is obtained.

[0020] Optionally, the entity relationships are represented in the disease knowledge graph as directed edges. The method for determining disease teaching relationships further includes:

[0021] The target teaching disease relationships are added to the disease knowledge graph in the form of directed edges.

[0022] Optional methods for determining disease-based teaching relationships include:

[0023] Obtain the features to be taught;

[0024] The features to be taught and the features corresponding to the teaching relationship of the target disease are compared to obtain the comparison results;

[0025] If the comparison results meet the similarity conditions, the teaching relationship between the feature to be taught and the target disease will be used as associated teaching materials.

[0026] A second aspect of this application provides a device for determining disease teaching relationships, comprising:

[0027] A construction unit is used to construct a disease knowledge graph, which includes entities and entity relationships between entities. The entities include disease type entities, feature type entities, and treatment type entities. The treatment type entities include core treatment type entities.

[0028] The first determining unit is used to determine multiple disease relationships corresponding to multiple historical case data based on the disease knowledge graph, wherein the disease relationship refers to the relationship between disease, feature and core treatment.

[0029] The second determining unit is used to determine multiple target disease relationships corresponding to the target core treatment from multiple disease relationships;

[0030] The third determining unit is used to determine multiple feature correlation degrees corresponding to multiple target disease relationships, wherein the feature correlation degree is used to represent the correlation between the feature corresponding to the target disease relationship and the target core treatment;

[0031] The setting unit is used to, if the feature correlation meets a preset correlation threshold, use the target disease relationship corresponding to the feature correlation as the target disease teaching relationship for the target core treatment.

[0032] A third aspect of this application provides an electronic device, including:

[0033] Memory, used to store executable instructions;

[0034] The processor, when executing executable instructions stored in the memory, implements the disease teaching relationship determination method provided in the first aspect of the embodiments of this application.

[0035] A fourth aspect of this application provides a computer-readable medium storing executable instructions, which, when executed by a processor, implement the disease teaching relationship determination method provided in the first aspect of this application.

[0036] In summary, this application provides a method and related apparatus for determining disease teaching relationships. The method includes: constructing a disease knowledge graph, which includes entities and entity relationships between entities. Entities include disease type entities, feature type entities, and treatment type entities. Treatment type entities include core treatment type entities. Based on the disease knowledge graph, multiple disease relationships corresponding to multiple historical case data are determined. A disease relationship refers to the relationship between a disease, a feature, and a core treatment. Multiple target disease relationships corresponding to a target core treatment are determined from the multiple disease relationships. Multiple feature relevances corresponding to the multiple target disease relationships are determined. Feature relevances represent the relevance between the feature corresponding to the target disease relationship and the target core treatment. If the feature relevance meets a preset relevance threshold, the target disease relationship corresponding to the feature relevance is taken as the target disease teaching relationship for the target core treatment. Through the above method, target teaching disease relationships for the target core treatment can be obtained, enabling the teaching of related diseases and features for the target core treatment. Attached Figure Description

[0037] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of this application. For those skilled in the art, other drawings can be obtained based on the provided drawings without creative effort.

[0038] Figure 1 A flowchart illustrating a method for determining disease teaching relationships provided in this application embodiment;

[0039] Figure 2 A detailed flowchart of a method for determining disease teaching relationships provided in an embodiment of this application;

[0040] Figure 3 A schematic diagram of a disease teaching relationship determination device provided in an embodiment of this application;

[0041] Figure 4 This is a schematic diagram of an electronic device provided in an embodiment of this application. Detailed Implementation

[0042] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present application, and not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present application.

[0043] The terms “first,” “second,” “third,” “fourth,” etc. (if present) in the specification, claims, and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a particular order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms “comprising” and “having,” and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0044] Existing disease teaching data mainly includes the relationship between diseases and characteristics, which can be used to train the part of clinical thinking from analyzing symptoms to judging diseases. However, existing disease teaching data only includes the part from analyzing symptoms to judging diseases, but does not include the part from judging diseases to giving treatment plans. In other words, existing disease teaching data cannot be used to teach treatment plans.

[0045] In view of this, embodiments of this application provide a method and related apparatus for determining disease teaching relationships, which can obtain target teaching disease relationships for target core treatments, so as to enable teaching of related diseases and characteristics for target core treatments.

[0046] The following is an explanation of the terms used in this application:

[0047] Core Management: While different measures are involved in the treatment of each disease, they can generally be categorized into two main types: medication and surgical procedures. Medications include adjuvant drugs and some symptomatic treatments (non-core management drugs), such as saline solution and antipyretics / analgesics. Surgical procedures are mostly related to the core management of the disease, especially the surgery itself, but some procedures are not core, such as intravenous catheterization and bladder irrigation. Therefore, core management refers to the most critical management measure for each disease diagnosis. For example, the core management for myocardial infarction patients is aspirin and coronary stent placement. Each core management corresponds to a higher-level category; for example, aspirin corresponds to antiplatelet drugs, and coronary stent placement corresponds to interventional reperfusion surgery. These are considered the core management categories for that disease, and their combinations are called core management category groups.

[0048] Disease phenotype: Disease phenotypes tend to be similar in different patients with the same disease, and they have a more definite correlation with treatment.

[0049] Treatment-level disease phenotype: The disease phenotype reflected by patients with the same diagnosis and the same treatment regimen is the treatment-level disease phenotype.

[0050] Evidence-based knowledge pathways: Evidence-based knowledge pathways are path-based disease representation systems that cover the initial characteristics of a disease to the final treatment measures. They include both medical theoretical data and real-world evidence-based information. Primary evidence-based knowledge pathways only involve information such as disease characteristic judgment and treatment classification.

[0051] Evidence-based medicine refers to the use of prospective, randomized, double-blind, controlled, and multicenter studies in clinical research to systematically collect and organize objective evidence obtained from large-sample studies as the basis for medical decisions.

[0052] Clinical guidelines, also known as medical guidelines or medical practice guidelines, are standards developed for the diagnosis, management, and treatment of clinical diseases and specialties, and serve as the basis for regulating clinical activities.

[0053] Expert consensus: This typically involves a team of experts from multiple disciplines reaching a consensus on the diagnosis and treatment of specific clinical problems, and it plays an important role in the development of guidelines.

[0054] Treatment plan: Based on the disease as a unit, and guided by evidence-based medicine, the treatment plan is formulated according to the individualized manifestations of the patient's symptoms, signs, tests, examinations, etc.

[0055] Treatment strategy: A collection of treatment options that can achieve the treatment goal; the selection of treatment options is guided by the treatment strategy.

[0056] Treatment measures: Specific medications and invasive procedures under the guidance of the treatment plan constitute the specific treatment measures.

[0057] Level of evidence: In clinical practice guidelines, if multiple large-scale prospective double-blind controlled studies have reached consistent conclusions for a certain treatment measure, the level of evidence is the highest and it is often listed as a strong recommendation; if there is no evidence-based medicine evidence and it is only logical reasoning, but it has been accepted by clinical practice, the level of evidence is the lowest and it is often listed as expert consensus or clinical treatment reference.

[0058] Machine learning: Machine learning is the process of using a computer to learn from a portion of data and then predict and judge other data. The core of machine learning is "using algorithms to analyze data, learn from it, and then make decisions or predictions about new data." In other words, it's a method where a computer uses acquired data to derive a model and then uses that model to make predictions.

[0059] Knowledge graphs, also known as knowledge domain visualizations or knowledge domain mapping maps, are a series of different graphics that display the development process and structural relationships of knowledge. They use visualization techniques to describe knowledge resources and their carriers, and to mine, analyze, construct, draw, and display knowledge and the interrelationships between them.

[0060] The following describes a method for determining disease teaching relationships provided in this application through method embodiments, such as... Figure 1 As shown, the method includes:

[0061] S101. Construct a disease knowledge graph.

[0062] A disease knowledge graph includes entities and the relationships between them. Entities in the disease knowledge graph are divided into three categories: disease type entities, feature type entities, and treatment type entities. Disease type entities refer to the name of the disease; feature type entities refer to the characteristics that a disease may have, including symptoms, signs, tests, examinations, family history, contraindications, and relevant medical history; treatment type entities refer to the treatments corresponding to the disease, including medications, invasive procedures, and surgeries. Among these, treatment type entities include core treatment type entities, which can be identified by labels. For example, core treatment type entities may include medications and / or surgeries. In addition, treatment type entities may also include non-core treatment type entities, such as adjuvant drugs.

[0063] Disease knowledge graphs can visualize three types of entities: disease type entities, feature type entities, and treatment type entities. For example, a disease knowledge graph can be represented by a directed graph G, as shown in the following formula:

[0064] G = (V, E)

[0065] In the formula, G represents the directed graph corresponding to the disease knowledge graph, V represents the set of points corresponding to entities, and E represents the set of edges corresponding to entity relations.

[0066] The point set corresponding to an entity can be represented by the following formula:

[0067] V = {1, 2, 3, ..., n}

[0068] In the formula, V represents the set of points corresponding to the entity, and n represents the number of entities.

[0069] The edge set corresponding to an entity relationship can be represented by the following formula:

[0070] E={(1,2),(2,3)……(n-1,n)}

[0071] In the formula, E represents the edge set corresponding to the entity relationship, (1,2) represents the relationship between entity 1 and entity 2, (2,3) represents the relationship between entity 2 and entity 3, and (n-1,n) represents the relationship between entity n-1 and entity n.

[0072] In a disease knowledge graph, each edge (u,v) is a directed edge from u to v, where u is the starting point and v is the ending point. For example, in a disease knowledge graph, (chest tightness, tracheal dilation) can represent the entity relationship between the feature type entity of chest tightness and the treatment type entity of tracheal dilation.

[0073] Disease knowledge graphs can visualize relevant information, which is more conducive to teaching clinical thinking and can also effectively improve the efficiency of information processing and transmission.

[0074] In one possible implementation, evidence-based medicine data can be obtained, which refers to data used in evidence-based medicine.

[0075] The evidence-based medicine data was segmented to obtain multiple evidence-based medicine terms.

[0076] A disease knowledge graph was obtained based on multiple evidence-based medicine terms.

[0077] Evidence-based medicine refers to data that can be used in evidence-based medicine. In the specific application of this embodiment, evidence-based medicine may include medical textbooks, literature, clinical guidelines, and expert consensus.

[0078] Before performing word segmentation on evidence-based medicine data, data cleaning and preprocessing can be performed on the evidence-based medicine data. Data cleaning refers to removing data that is irrelevant to the disease, characteristics, and treatment, while data preprocessing refers to processing data with different formats into the same format.

[0079] Then, the cleaned and preprocessed evidence-based medicine data is segmented to obtain multiple evidence-based medicine terms. Finally, a disease knowledge graph is obtained based on these multiple evidence-based medicine terms.

[0080] In the era of big data, although it is easy to obtain domestic and international medical information, integrating the materials still requires a lot of time and effort. Therefore, in this embodiment, a large amount of evidence-based medicine is integrated, and the data is cleaned and preprocessed to form a disease knowledge graph. The entities and entity relationships in the disease knowledge graph can be better processed and are more convenient to be applied in the field of teaching.

[0081] S102. Based on the disease knowledge graph, determine the relationships between multiple diseases corresponding to multiple historical case data.

[0082] Historical case data refers to actual case data, while disease relationships refer to the relationships between diseases, characteristics, and core treatments.

[0083] Since a disease knowledge graph includes entity relationships between disease type entities, feature type entities, and treatment type entities, the disease relationships corresponding to historical case data can be determined based on the disease knowledge graph. In other words, the relationships between diseases, features, and core treatments reflected in historical case data can be determined.

[0084] It should be noted that, since historical case data are real case data, their corresponding characteristics are more individualized and diverse, which can make the obtained disease relationships more comprehensive.

[0085] S103. Identify multiple target disease relationships corresponding to the core treatment from multiple disease relationships.

[0086] In this embodiment, multiple target disease relationships corresponding to the target core treatment can be determined from multiple disease relationships. The core treatment corresponding to the target disease relationship is the same as the target core treatment. That is, multiple target disease relationships related to the target core treatment are selected from multiple disease relationships. Since the disease phenotype reflected by patients with the same treatment plan under the same diagnosis is the treatment level disease phenotype, multiple features in multiple target disease relationships for the target core treatment form a set of disease phenotypes with the same treatment level.

[0087] It should be noted that in this embodiment, there may be multiple target core treatments. When there are multiple target core treatments, multiple disease relationships can be screened separately for these multiple target core treatments, thereby determining the multiple target disease relationship groups corresponding to the multiple target core treatments.

[0088] S104. Determine the correlation of multiple features corresponding to multiple target disease relationships.

[0089] Feature relevance is used to represent the correlation between the features corresponding to the target disease and the target core treatment.

[0090] Since the amount of disease relationship data is usually quite large, there may be a situation where the feature corresponding to a certain target disease relationship has a low correlation with the target core treatment among multiple target disease relationships. In order to improve accuracy, multiple feature correlations corresponding to multiple target disease relationships can be determined. Feature correlation is used to represent the correlation between the feature corresponding to the target disease relationship and the target core treatment.

[0091] In one possible implementation, S104 determines multiple feature correlations corresponding to multiple target disease relationships, including:

[0092] For the i-th target disease relationship among multiple target disease relationships, obtain the feature probability of having a feature corresponding to the i-th target disease relationship, the target core treatment probability of taking the target core treatment, and the target core treatment conditional probability of taking the target core treatment when a feature corresponding to the i-th target disease relationship exists;

[0093] Input the feature probability, the target core treatment probability, and the target core treatment conditional probability into the following formula to obtain the feature relevance corresponding to the i-th target disease relationship:

[0094]

[0095] Among them, P i P(A) refers to the feature relevance corresponding to the i-th target disease relationship. i P(B) refers to the feature probability that the i-th target disease relationship corresponds to a feature, and P(B) refers to the target core treatment probability. i ) refers to the conditional probability of taking the target core treatment when there is a corresponding feature of the i-th target disease relationship, where i is a positive integer.

[0096] Using the above formula, for the i-th target disease relationship among multiple target disease relationships, the correlation between the corresponding feature and the target core treatment can be accurately determined, that is, the feature correlation corresponding to the i-th target disease relationship is determined. The higher the value of the feature correlation corresponding to the i-th target disease relationship, the greater the probability that the feature corresponding to the i-th target disease relationship and the target core treatment are the correct combination of disease, feature and core treatment.

[0097] It should be noted that since the i-th target disease relationship refers to any one of the multiple target disease relationships, the feature relevance of each target disease relationship in the multiple target disease relationships can be determined by referring to the above method.

[0098] S105. If the feature correlation meets the preset correlation threshold, the target disease relationship corresponding to the feature correlation is used as the target disease teaching relationship for the target core treatment.

[0099] If the feature relevance meets the preset relevance threshold, where the preset relevance threshold can be determined in advance by relevant personnel based on experience, it indicates that the feature in the target disease relationship corresponding to the feature relevance has a high degree of relevance to the target core treatment. Therefore, the target disease relationship corresponding to the feature relevance can be used as the target disease teaching relationship for the target core treatment, so that relevant diseases and features can be taught according to the target disease teaching relationship for the target core treatment.

[0100] Since the clinical diagnosis and treatment logic proceeds from analyzing symptoms to determining the disease and then to giving a treatment plan, the target disease teaching relationship determined in this embodiment includes the disease, characteristics, and target core treatment. In this way, teaching can be conducted for analyzing symptoms, determining the disease, and giving a treatment plan in clinical thinking.

[0101] In one possible implementation, in order to visualize the relationships between the target teaching diseases, when the entity relationships are represented as directed edges in the disease knowledge graph, the target teaching disease relationships can be added to the disease knowledge graph as directed edges.

[0102] Specifically, a directed edge can be generated by using the disease as the vertex and the target core treatment as the endpoint in the target teaching disease relationship. Features in the target teaching disease relationship can be located on this directed edge. By adding the determined target teaching disease relationship to the disease knowledge graph in the form of directed edges, the determined target teaching disease relationship can be supplemented into the disease knowledge graph. Furthermore, the directed edges enable a visual display of the target teaching disease relationship, facilitating teaching based on the target teaching disease relationship.

[0103] It should be noted that there may be multiple target teaching disease relationships for target core treatment and disease. Therefore, in order to display these multiple target teaching disease relationships in a unified manner, the features corresponding to the target teaching disease relationships can be arranged sequentially on the directed edges corresponding to the disease and the target core treatment.

[0104] In one possible implementation, the features to be taught can be obtained;

[0105] The characteristics corresponding to the teaching features and the teaching relationship of the target disease were compared to obtain the comparison results.

[0106] If the comparison results meet the similarity criteria, the teaching relationship between the teaching feature to be taught and the teaching relationship of the target disease will be used as related teaching materials.

[0107] In the application of teaching relationships for target diseases, the features to be taught and the features corresponding to the teaching relationships for target diseases can be compared to obtain comparison results. The comparison results can be used to indicate the degree of similarity between the features to be taught and the features corresponding to the teaching relationships for target diseases. If the comparison results meet the similarity conditions, it means that the degree of similarity between the features to be taught and the features corresponding to the teaching relationships for target diseases has reached the preset similarity threshold. The preset similarity threshold can be set by relevant staff based on experience. This indicates that the degree of similarity between the features to be taught and the features corresponding to the teaching relationships for target diseases is high. Therefore, the features to be taught and the teaching relationships for target diseases can be used as related teaching materials, thereby enriching the relevant teaching materials for the core treatment of the target disease.

[0108] The following is based on Figure 2This application will explain the specific process of determining the disease-teaching relationship provided, such as... Figure 2 As shown, the specific process includes:

[0109] First, evidence-based medicine (EBR) data about the disease is obtained. Then, the data is processed, such as cleaning and preprocessing to address formatting issues. Next, word segmentation is performed to obtain EBR terms related to the disease, treatment strategies, core treatments, patient characteristics, and treatment measures. Finally, a knowledge graph (disease knowledge graph) is constructed based on these EBR terms. The disease knowledge graph includes entities and the relationships between entities.

[0110] Then, using real case data as historical case data, multiple disease relationships are identified. These relationships include those between diseases, patient characteristics (features), and core treatment items (core treatments). Since there are usually multiple disease relationships for a single core treatment item, a directed edge can be constructed with the disease as the starting point and the core treatment item as the ending point. The patient characteristics corresponding to these multiple disease relationships are then arranged sequentially on this directed edge. To ensure data reliability, the weights (feature relevance) of the patient characteristics corresponding to the disease relationships can be calculated using a formula. This determines the relevance between the patient characteristics and the core treatments corresponding to the disease relationships, and the reliability of the disease relationships is verified based on this relevance. Only disease relationships with high reliability can be identified as teaching disease relationships.

[0111] Finally, once the relationships between multiple teaching diseases for the core treatment items have been identified, in order to achieve disease prediction during the teaching process, new patient characteristics can be used as teaching characteristics, compared with known patient characteristics, and the comparison results can be used to determine whether the teaching characteristics and these multiple teaching disease relationships are related teaching materials.

[0112] Therefore, this application provides a method and related apparatus for determining disease teaching relationships. The method includes: constructing a disease knowledge graph, which includes entities and entity relationships between entities. Entities include disease type entities, feature type entities, and treatment type entities. Treatment type entities include core treatment type entities. Based on the disease knowledge graph, multiple disease relationships corresponding to multiple historical case data are determined. A disease relationship refers to the relationship between a disease, a feature, and a core treatment. Multiple target disease relationships corresponding to a target core treatment are determined from the multiple disease relationships. Multiple feature relevances corresponding to the multiple target disease relationships are determined. Feature relevances represent the relevance between the feature corresponding to the target disease relationship and the target core treatment. If the feature relevance meets a preset relevance threshold, the target disease relationship corresponding to the feature relevance is taken as the target disease teaching relationship for the target core treatment. Through the above method, target teaching disease relationships for the target core treatment can be obtained, so that related diseases and features can be taught for the target core treatment.

[0113] In the foregoing Figure 1-2 Based on the corresponding embodiments, Figure 3 This is a schematic diagram of a disease teaching relationship determination device provided in an embodiment of this application. The disease teaching relationship determination device 300 includes:

[0114] Construction unit 301 is used to construct a disease knowledge graph, the disease knowledge graph includes entities and entity relationships between entities, the entities include disease type entities, feature type entities and treatment type entities, the treatment type entities include core treatment type entities;

[0115] The first determining unit 302 is used to determine multiple disease relationships corresponding to multiple historical case data based on the disease knowledge graph, wherein the disease relationship refers to the relationship between disease, feature and core treatment.

[0116] The second determining unit 303 is used to determine multiple target disease relationships corresponding to the target core treatment from multiple disease relationships;

[0117] The third determining unit 304 is used to determine multiple feature correlation degrees corresponding to multiple target disease relationships, wherein the feature correlation degree is used to represent the correlation between the feature corresponding to the target disease relationship and the target core treatment;

[0118] Setting unit 305 is used to, if the feature correlation meets a preset correlation threshold, use the target disease relationship corresponding to the feature correlation as the target disease teaching relationship for the target core treatment.

[0119] In one possible implementation, the third determining unit is used for:

[0120] For the i-th target disease relationship among multiple target disease relationships, obtain the feature probability of the i-th target disease relationship having a corresponding feature, the target core treatment probability of taking the target core treatment, and the target core treatment conditional probability of taking the target core treatment when the i-th target disease relationship has a corresponding feature;

[0121] By inputting the feature probability, the target core treatment probability, and the target core treatment conditional probability into the following formula, the feature relevance corresponding to the i-th target disease relationship is obtained:

[0122]

[0123] Among them, P i P(A) refers to the feature relevance corresponding to the i-th target disease relationship. i P(B) refers to the feature probability of the i-th target disease relationship corresponding to the feature, and P(B|A) refers to the target core treatment probability of taking the target core treatment. i ) refers to the conditional probability of taking the target core treatment when there is an i-th feature corresponding to the target disease relationship, where i is a positive integer.

[0124] In one possible implementation, the disease teaching relationship determination device 300 further includes an acquisition unit for:

[0125] Acquire evidence-based medicine data, which refers to data used in evidence-based medicine;

[0126] The evidence-based medicine data was segmented to obtain multiple evidence-based medicine terms;

[0127] Based on multiple evidence-based medicine terms, a disease knowledge graph is obtained.

[0128] In one possible implementation, the disease teaching relationship determination device 300 further includes an adding unit for:

[0129] If the entity relationship is represented as a directed edge in the disease knowledge graph, the target teaching disease relationship is added to the disease knowledge graph as a directed edge.

[0130] In one possible implementation, the disease teaching relationship determination device 300 further includes an acquisition unit for:

[0131] Obtain the features to be taught;

[0132] The features to be taught and the features corresponding to the teaching relationship of the target disease are compared to obtain the comparison results;

[0133] If the comparison results meet the similarity conditions, the teaching relationship between the feature to be taught and the target disease will be used as associated teaching materials.

[0134] It should be noted that the specific working process of each module provided in the above embodiments of this application can be referred to the corresponding implementation methods in the above method embodiments, and will not be repeated here.

[0135] Therefore, the disease teaching relationship determination device is used to: construct a disease knowledge graph, which includes entities and the relationships between them. Entities include disease type entities, feature type entities, and treatment type entities, with treatment type entities including core treatment type entities; determine multiple disease relationships corresponding to multiple historical case data based on the disease knowledge graph, where a disease relationship refers to the relationship between a disease, a feature, and a core treatment; determine multiple target disease relationships corresponding to the target core treatment from these multiple disease relationships; determine multiple feature relevances corresponding to each of the multiple target disease relationships, where feature relevance represents the relevance between the feature corresponding to the target disease relationship and the target core treatment; if the feature relevance meets a preset relevance threshold, the target disease relationship corresponding to the feature relevance is used as the target disease teaching relationship for the target core treatment. Through this device, target teaching disease relationships for the target core treatment can be obtained, enabling the teaching of related diseases and features for the target core treatment.

[0136] Another embodiment of this application provides an electronic device, such as... Figure 4 As shown, it includes:

[0137] Memory, used to store executable instructions;

[0138] A processor, when executing executable instructions stored in memory, implements the method described in the embodiments of this application.

[0139] Another embodiment of this application provides a computer-readable storage medium storing executable instructions for implementing the methods described above in the embodiments of this application when executed by a processor.

[0140] Those skilled in the art will further recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of both. To clearly illustrate the interchangeability of hardware and software, the components and steps of the various examples have been generally described in terms of functionality in the foregoing description. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0141] The above description of the disclosed embodiments enables those skilled in the art to make or use this application. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of this application. Therefore, this application is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for determining disease teaching relationships, characterized in that, The method includes: A disease knowledge graph is constructed based on multiple evidence-based medicine terms. The disease knowledge graph includes entities and the entity relationships between entities. The entities include disease type entities, feature type entities, and treatment type entities. The disease type entity refers to the name of the disease, the feature type entity refers to the characteristics that the disease may have, and the treatment type entity refers to the treatment corresponding to the disease. The treatment type entities include core treatment type entities and non-core treatment type entities. The core treatment type entities include drugs and / or surgeries, and the non-core treatment type entities include adjuvant drugs. Based on the disease knowledge graph, multiple disease relationships are determined corresponding to multiple historical case data. The disease relationship refers to the relationship between diseases, characteristics and core treatments in multiple historical case data. Identify multiple target disease relationships corresponding to the target core treatment from among the multiple disease relationships described above; For the i-th target disease relationship among the multiple target disease relationships, obtain the feature probability of the i-th target disease relationship having a corresponding feature, the target core treatment probability of taking the target core treatment, and the target core treatment conditional probability of taking the target core treatment when the i-th target disease relationship has a corresponding feature; Based on the feature probability, the target core treatment probability, and the target core treatment conditional probability, the feature correlation degree corresponding to the i-th target disease relationship is obtained, and the feature correlation degree is used to represent the correlation between the feature corresponding to the target disease relationship and the target core treatment; If the feature correlation meets the preset correlation threshold, the target disease relationship corresponding to the feature correlation is used as the target disease teaching relationship for the target core treatment.

2. The method according to claim 1, characterized in that, The determination of multiple feature correlations corresponding to the multiple target disease relationships includes: By inputting the feature probability, the target core treatment probability, and the target core treatment conditional probability into the following formula, the feature relevance corresponding to the i-th target disease relationship is obtained: ； in, This refers to the feature relevance corresponding to the i-th target disease relationship. It refers to the feature probability that the i-th target disease relationship corresponds to a certain feature. This refers to the probability of taking the target core action. It refers to the conditional probability of taking the target core treatment when there is an i-th feature corresponding to the target disease relationship, where i is a positive integer.

3. The method according to claim 1, characterized in that, The method further includes: Acquire evidence-based medicine data, which refers to data used in evidence-based medicine; The evidence-based medicine data was segmented to obtain multiple evidence-based medicine terms; Based on multiple evidence-based medicine terms, a disease knowledge graph is obtained.

4. The method according to claim 1, characterized in that, The entity relationships are represented in the form of directed edges in the disease knowledge graph, and the method further includes: The target teaching disease relationships are added to the disease knowledge graph in the form of directed edges.

5. The method according to any one of claims 1-4, characterized in that, The method also includes Obtain the features to be taught; The features to be taught and the features corresponding to the teaching relationship of the target disease are compared to obtain the comparison results; If the comparison results meet the similarity conditions, the teaching relationship between the feature to be taught and the target disease will be used as associated teaching materials.

6. A device for determining disease teaching relationships, characterized in that, The device includes: A construction unit is used to construct a disease knowledge graph based on multiple evidence-based medicine terms. The disease knowledge graph includes entities and entity relationships between entities. The entities include disease type entities, feature type entities, and treatment type entities. The disease type entity refers to the name of the disease, the feature type entity refers to the characteristics that the disease may have, and the treatment type entity refers to the treatment corresponding to the disease. The treatment type entities include core treatment type entities and non-core treatment type entities. The core treatment type entities include drugs and / or surgeries, and the non-core treatment type entities include adjuvant drugs. The first determining unit is used to determine multiple disease relationships corresponding to multiple historical case data based on the disease knowledge graph. The disease relationship refers to the relationship between diseases, characteristics and core treatments in multiple historical case data. The second determining unit is used to determine multiple target disease relationships corresponding to the target core treatment from multiple disease relationships; The third determining unit is configured to, for the i-th target disease relationship among the plurality of target disease relationships, obtain the feature probability of the existence of the feature corresponding to the i-th target disease relationship, the target core treatment probability of taking the target core treatment, and the target core treatment conditional probability of taking the target core treatment when the feature corresponding to the i-th target disease relationship exists; and based on the feature probability, the target core treatment probability, and the target core treatment conditional probability, obtain the feature relevance corresponding to the i-th target disease relationship, wherein the feature relevance is used to represent the relevance between the feature corresponding to the target disease relationship and the target core treatment; The setting unit is used to, if the feature correlation meets a preset correlation threshold, use the target disease relationship corresponding to the feature correlation as the target disease teaching relationship for the target core treatment.

7. The apparatus according to claim 6, characterized in that, The third determining unit is used for: By inputting the feature probability, the target core treatment probability, and the target core treatment conditional probability into the following formula, the feature relevance corresponding to the i-th target disease relationship is obtained: ； in, This refers to the feature relevance corresponding to the i-th target disease relationship. It refers to the feature probability that the i-th target disease relationship corresponds to a certain feature. This refers to the probability of taking the target core action. It refers to the conditional probability of taking the target core treatment when there is an i-th feature corresponding to the target disease relationship, where i is a positive integer.

8. The apparatus according to claim 6, characterized in that, The device further includes an acquisition unit for: Obtain the features to be taught; The features to be taught and the features corresponding to the teaching relationship of the target disease are compared to obtain the comparison results; If the comparison results meet the similarity conditions, the teaching relationship between the feature to be taught and the target disease will be used as associated teaching materials.

9. An electronic device, characterized in that, include: Memory, used to store executable instructions; The processor, when executing executable instructions stored in the memory, implements the method for determining disease teaching relationships as described in any one of claims 1 to 5.

10. A computer-readable storage medium, characterized in that, It stores executable instructions for implementing the disease teaching relationship determination method according to any one of claims 1 to 5 when executed by a processor.

Images