Medical information display method and device, computer device and storage medium

Abstract

This application discloses a method, apparatus, computer device, and storage medium for displaying medical information. The method includes: displaying a medical information query interface including a query object input box and a query result display area; and displaying medical information query results matching a target query object in the query result display area, wherein the target query object is the query object entered in the query object input box. When the medical information query interface is a medical knowledge graph query interface, the target query object is the medical entity to be queried, and the medical information query result is a matching medical knowledge graph that matches the medical entity to be queried. When the medical information query interface is a medical ontology graph query interface, the target query object is the medical concept object to be queried, and the medical information query result is a matching medical ontology graph that matches the medical concept object to be queried. This method enables the querying of medical information and enriches the types of medical information available for querying.

Description

Technical Field

[0001] This application relates to the field of computer technology, and in particular to medical information display methods, medical information display devices, computer equipment, and computer-readable storage media. Background Technology

[0002] Medical information plays a crucial guiding role in the work of healthcare professionals. For example, medical knowledge graphs, which are semantic networks revealing the relationships between medical entities, provide a formal description of real-world medical-related things and their relationships. Healthcare professionals can quickly understand the connections between medical entities using medical knowledge graphs. For instance, in a medical knowledge graph centered on "sulbactam sodium," "sulbactam sodium" and "cholecystitis" are related: the "disease drug" for "cholecystitis" is "sulbactam sodium." Healthcare professionals can then proceed with subsequent medical procedures based on this relationship. Clearly, medical knowledge graphs are extremely convenient and important for healthcare professionals.

[0003] Therefore, it is essential to know how to search for medical information. Summary of the Invention

[0004] This application provides a method, apparatus, computer equipment, and storage medium for displaying medical information, which can enable the querying of medical information and enrich the types of medical information available for querying.

[0005] One embodiment of this application discloses a method for displaying medical information, the method comprising:

[0006] The medical information query interface includes a query object input box and a query result display area.

[0007] The query results display area displays medical information query results that match the target query object, which is the query object entered in the query object input box;

[0008] The medical information query interface can be either a medical knowledge graph query interface or a medical ontology graph query interface. When the medical information query interface is a medical knowledge graph query interface, the target query object is the medical entity to be queried, and the medical information query result is a matching medical knowledge graph that matches the medical entity to be queried. When the medical information query interface is a medical ontology graph query interface, the target query object is the medical concept object to be queried, and the medical information query result is a matching medical ontology graph that matches the medical concept object to be queried.

[0009] One embodiment of this application discloses a medical information display device, which includes:

[0010] The display unit is used to display a medical information query interface, which includes a query object input box and a query result display area.

[0011] The display unit is also used to display medical information query results that match the target query object in the query result display area, wherein the target query object is the query object entered in the query object input box;

[0012] The medical information query interface can be either a medical knowledge graph query interface or a medical ontology graph query interface. When the medical information query interface is a medical knowledge graph query interface, the target query object is the medical entity to be queried, and the medical information query result is a matching medical knowledge graph that matches the medical entity to be queried. When the medical information query interface is a medical ontology graph query interface, the target query object is the medical concept object to be queried, and the medical information query result is a matching medical ontology graph that matches the medical concept object to be queried.

[0013] One aspect of this application discloses a computer device, which includes a processor adapted to implement one or more computer programs; and a computer storage medium storing one or more computer programs, the one or more computer programs being adapted to be loaded and executed by the processor for the above-described medical information display method.

[0014] One aspect of this application discloses a computer-readable storage medium storing one or more computer programs adapted to be loaded by a processor and executed by the above-described medical information display method.

[0015] One embodiment of this application discloses a computer program product, which includes a computer program stored in a computer-readable storage medium. A processor of a computer device reads the computer program from the computer-readable storage medium and executes the computer program, causing the computer device to perform the aforementioned medical information display method.

[0016] This application embodiment displays a medical information query interface, which includes a query object input box and a query result display box. When a target query object is entered in the query object input box, the medical information query results matching the target query object are displayed in the query result display area. The medical information query interface can be a medical knowledge graph query interface or a medical ontology graph query interface. When the medical information query interface is a medical knowledge graph query interface, the target query object is the medical entity to be queried, and the medical information query results are matching medical knowledge graphs that match the medical entity to be queried. Based on this medical knowledge graph query interface, various medical entities can be queried, enabling medical workers to quickly grasp the medical knowledge graphs corresponding to each medical entity. When the medical information query interface is a medical ontology graph query interface, the target query object is the medical concept object to be queried, and the medical information query results are matching medical ontology graphs that match the medical concept object to be queried. Based on this medical ontology graph query interface, various medical concept objects can be queried, helping medical workers quickly understand the current medical ontology and quickly locate high-frequency important medical concepts and meta-relationships. Therefore, the medical information display method proposed in this application can realize the query of medical information and enrich the types of medical information available for query. Attached Figure Description

[0017] 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 some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1 This is a schematic diagram of the network architecture of a medical information display system disclosed in an embodiment of this application;

[0019] Figure 2 This is a flowchart illustrating a medical information display method disclosed in an embodiment of this application;

[0020] Figure 3 This is a schematic diagram of a medical information query interface disclosed in an embodiment of this application;

[0021] Figure 4 This is a schematic diagram of another medical information query interface disclosed in the embodiments of this application;

[0022] Figure 5 This is a flowchart illustrating another medical information display method disclosed in an embodiment of this application;

[0023] Figure 6This is a schematic diagram of a medical knowledge graph query interface disclosed in an embodiment of this application;

[0024] Figure 7 This is a schematic diagram of another medical knowledge graph query interface disclosed in the embodiments of this application;

[0025] Figure 8 This is a flowchart illustrating another medical information display method disclosed in the embodiments of this application;

[0026] Figure 9 This is a schematic diagram of a medical ontology atlas query interface disclosed in an embodiment of this application;

[0027] Figure 10 This is a schematic diagram of another medical ontology atlas query interface disclosed in the embodiments of this application;

[0028] Figure 11 This is a schematic diagram of another medical ontology atlas query interface disclosed in the embodiments of this application;

[0029] Figure 12 This is an interface diagram of a medical information display system disclosed in an embodiment of this application;

[0030] Figure 13 This is a schematic diagram of a medical information statistics interface disclosed in an embodiment of this application;

[0031] Figure 14 This is a schematic diagram of a backend technology framework disclosed in an embodiment of this application;

[0032] Figure 15 This is a schematic diagram of a recall module for inverted indexing and co-occurrence calculation disclosed in an embodiment of this application;

[0033] Figure 16 This is a model architecture diagram of a medical entity matching stage disclosed in an embodiment of this application;

[0034] Figure 17 This is a schematic diagram of the structure of a medical information display device disclosed in an embodiment of this application;

[0035] Figure 18 This is a schematic diagram of the structure of a computer device disclosed in an embodiment of this application. Detailed Implementation

[0036] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0037] To enable richer medical information retrieval, this application proposes a medical information display method. Based on this method, a medical knowledge graph query interface and a medical ontology graph query interface can be displayed, allowing users to query a wealth of medical information. The medical knowledge graph (also known as a medical knowledge graph) is essentially a semantic network revealing the relationships between medical entities. It can formally describe things in the real world and their interrelationships. The medical ontology graph (also known as a medical ontology) defines hierarchical categories and attribute information, such as the entity category "disease," and hierarchical structures from "treatment method" to "surgery." The medical information display method provided in this application can be implemented based on Artificial Intelligence (AI) technology. AI refers to the theory, methods, technologies, and application systems that use digital computers or machines controlled by digital computers to simulate, extend, and expand human intelligence, perceive the environment, acquire knowledge, and use knowledge to obtain optimal results. AI technology is a comprehensive discipline involving a wide range of fields; the medical information display method provided in this application mainly involves Machine Learning (ML) technology within AI. Machine learning (ML) is a multidisciplinary field involving probability theory, statistics, approximation theory, convex analysis, and algorithm complexity theory. It specifically studies how computers can simulate or implement human learning behavior to acquire new knowledge or skills and reorganize existing knowledge structures to continuously improve their performance. Machine learning is the core of artificial intelligence and the fundamental way to endow computers with intelligence; its applications span all areas of artificial intelligence. Machine learning and deep learning typically include techniques such as artificial neural networks, belief networks, reinforcement learning, transfer learning, inductive learning, and instructional learning.

[0038] In feasible embodiments, the medical information display method provided in this application can also be implemented based on cloud technology and / or blockchain technology. Specifically, it may involve one or more of cloud technologies, such as cloud storage, cloud database, and big data. For example, data required to execute the medical information display method (e.g., medical information query results, specifically matching medical knowledge graphs that match the medical entity to be queried or matching medical ontology graphs that match the medical concept object to be queried) can be obtained from the cloud database. As another example, the data required to execute the medical information display method can be stored on the blockchain in the form of blocks; the data generated by executing the medical information display method (e.g., medical information query results, specifically matching medical knowledge graphs that match the medical entity to be queried or matching medical ontology graphs that match the medical concept object to be queried) can be stored on the blockchain in the form of blocks; furthermore, the data processing device (mainly referring to the backend technology) executing the medical information display method can be a node device in the blockchain network.

[0039] Please see Figure 1 , Figure 1 This is a schematic diagram of the network architecture of a medical information display system according to an embodiment of this application, such as... Figure 1 As shown, the medical information display system 100 may include at least a user 101 and a computer device 102. The computer device 102 is mainly used to execute the medical information display method, including displaying a medical information query interface and displaying medical information query results matching the target query object. The user 101 can be any medical professional or the administrator of the medical information system. Medical professionals can query the required medical information through the computer device 102, and administrators can update the medical information system through the computer device 102.

[0040] In one possible implementation, the computer device 102 mentioned above includes, but is not limited to, mobile phones, computers, intelligent voice interaction devices, smart home appliances, vehicle terminals, aircraft, etc.; wherein, the computer device 102 may also include a server, which may be an independent physical server, or a server cluster or distributed system composed of multiple physical servers, or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, CDN, and big data and artificial intelligence platforms. Figure 1 This is merely an illustrative network architecture diagram of a medical information display system and is not intended to be limiting. For example, Figure 1The computer device 102 includes a server that can be deployed as a node in a blockchain network, or the computer device 102 can be directly connected to the blockchain network. This allows the computer device 102 to upload the data generated when executing the medical information display method, such as medical information query results (matching medical knowledge graphs that match the medical entities to be queried and matching medical ontology graphs that match the medical concept objects to be queried), to the blockchain network for storage, so as to prevent internal data from being tampered with and thus ensure data security.

[0041] Based on the aforementioned medical information display system, the medical information display method of this application embodiment can generally include: displaying a medical information query interface, which includes a query object input box and a query result display box. When a target query object is entered in the query object input box, the medical information query results matching the target query object are displayed in the query result display area. The medical information query interface can be a medical knowledge graph query interface or a medical ontology graph query interface. When the medical information query interface is a medical knowledge graph query interface, the target query object is the medical entity to be queried, and the medical information query result is a matching medical knowledge graph that matches the medical entity to be queried. Based on this medical knowledge graph query interface, various medical entities can be queried, enabling medical workers to quickly grasp the medical knowledge graph corresponding to each medical entity. When the medical information query interface is a medical ontology graph query interface, the target query object is the medical concept object to be queried, and the medical information query result is a matching medical ontology graph that matches the medical concept object to be queried. Based on this medical ontology graph query interface, various medical concept objects can be queried, helping medical workers quickly understand the current medical ontology and quickly locate high-frequency important medical concept objects and meta-relationships. Therefore, the medical information display method proposed in this application can realize the query of medical information and enrich the types of medical information available for query.

[0042] The embodiments of this application are mainly applicable to medical scenarios, where medical knowledge graphs and medical ontology graphs can help users (mainly medical staff and administrators (administrators can refer to developers)) to understand structured medical knowledge to the greatest extent.

[0043] It is understood that, in the specific embodiments of this application, the data related to the matching medical knowledge graph and the medical concept object to be queried, which are matched with the medical entity to be queried, need to comply with the relevant laws, regulations and standards of the relevant countries and regions when the embodiments of this application are applied to specific products or technologies.

[0044] Based on the above description of the medical information display system, this application discloses a medical information display method. Please refer to [link to relevant documentation]. Figure 2This is a flowchart illustrating a medical information display method disclosed in an embodiment of this application. The medical information display method includes, but is not limited to, the following steps:

[0045] S201: Display the medical information query interface, which includes a query object input box and a query result display area.

[0046] In one possible implementation, when a user opens the medical information system, a medical information query interface is displayed. This interface includes a query input box and a query results display area. The medical information query interface can be either a medical knowledge graph query interface or a medical ontology graph query interface. For different medical information query interfaces, the format of the query input box can be the same or different. For example, the query input box can be a dropdown menu, allowing the user to select from it; or it can be a text input box, allowing the user to type text information. The content displayed in the query results display area also differs depending on the medical information query interface.

[0047] In this application, for the medical knowledge graph query interface, the corresponding query object input box can be entered by typing, because there are many types of medical entities, and it is more convenient for users to customize the input for querying; for the medical ontology graph query interface, the corresponding query object input box can be a drop-down list, because ontology displays the relationship between concepts, and multiple medical entities may correspond to one medical concept object. Therefore, there are fewer types of medical concept objects than medical entities, and using a drop-down list is more convenient and can also avoid errors (when one is not familiar with medical concept objects, one may enter incorrect expressions).

[0048] The following is an introduction to the medical information query interface, for example, such as... Figure 3 The diagram shown is a schematic of a medical information query interface disclosed in an embodiment of this application. The selected option is "Knowledge Graph." As can be seen, this interface includes an object input box 301 and a query result display area 302. Figure 3As can be seen, the medical entity "diabetes" was entered in the object input box 301, and correspondingly, a medical knowledge graph for "diabetes" is displayed in the query results display area 302. It can be seen that this medical knowledge graph is centered on "diabetes" and connects other medical entities related to "diabetes." For example, "diabetes" and "Xiaotangling Granules" are directly related: the "disease drug" corresponding to "diabetes" is "Xiaotangling Granules." Other relationships exist for the medical entity "diabetes," such as "primary disease" and "disease testing," which will not be elaborated here. Compared to current medical knowledge graphs, the medical knowledge graph displayed in this application is more detailed, with corresponding relationships between different medical entities, making it more readable and easier for users to understand.

[0049] For example, such as Figure 4 The diagram shown is a schematic of another medical information query interface disclosed in this application embodiment. Here, "Ontology Atlas" is selected. As can be seen, this interface includes an object input box 401 and a query result display area 402. Figure 4 As can be seen, the medical concept object "e_doctor" was entered into the object input box 401. Correspondingly, the medical ontology graph for "e_doctor" is displayed in the query results display area 402. The graph shows that this medical ontology graph is centered on "e_doctor," with other medical concept objects connected around it. For example, "e_doctor" is connected to "e_drug," and their corresponding relationship is "r_introduction." Figure 4 It can also be seen that there may be multiple relationships between two medical concept objects. For example, "e_doctor" and "e_disease" have two relationships: "r_specializes in diseases" and "r_introduction".

[0050] S202: Display medical information query results that match the target query object in the query results display area. The target query object is the query object entered in the query object input box.

[0051] In one possible implementation, when the medical information query interface is a medical knowledge graph query interface, the target query object is the medical entity to be queried, and the medical information query result is the matching medical knowledge graph that matches the medical entity to be queried. That is, when the medical information query interface is a medical knowledge graph query interface, if the input box for the query object in the medical knowledge graph query interface contains the medical entity to be queried, then the matching medical knowledge graph that matches the medical entity to be queried is displayed in the query results display area of ​​the medical knowledge graph query interface. For example, such as... Figure 3 As shown, "diabetes" is the medical entity to be queried, and the 302 error indicates the matching medical knowledge graph that matches "diabetes".

[0052] In another possible implementation, when the medical information query interface is a medical ontology graph query interface, the target query object is the medical concept object to be queried, and the medical information query result is the matching medical ontology graph that matches the medical concept object to be queried. That is, when the medical information query interface is a medical ontology graph query interface, if the input box for the query object in the medical ontology graph query interface contains the medical concept object to be queried, then the matching medical ontology graph that matches the medical concept object to be queried is displayed in the query results display area of ​​the medical ontology graph query interface. For example, ... Figure 4 As shown, “e_doctor” is the medical concept object to be queried, and 402 shows the matching medical ontology graph that matches “e_doctor”.

[0053] In this embodiment of the application, when querying medical information, a medical information query interface can be displayed first. Then, based on the user's operation, the medical information query results matching the target query object can be displayed in the query results display area. For different query interfaces, the displayed medical information query results will also be different depending on the query object entered in the query object input box. In addition to displaying a medical knowledge graph, this application can also display a medical ontology graph. Through multiple interfaces, users can query various types of medical information to obtain richer and more comprehensive medical information.

[0054] The above brief description of medical information query interfaces suggests that they can include medical knowledge graph query interfaces and medical ontology graph query interfaces. The displayed graph content and format will differ depending on the specific query interface. Please refer to [link / reference]. Figure 5 This is a flowchart illustrating another medical information display method disclosed in an embodiment of this application. Figure 5 The method for displaying medical information shown is illustrated using a medical information query interface as a medical knowledge graph query interface, and includes, but is not limited to, the following steps:

[0055] S501: Display the medical knowledge graph query interface, which includes a query object input box and a query result display area.

[0056] like Figure 6 As shown in 610, when a user clicks "Knowledge Graph," a medical knowledge graph query interface will be displayed, including a query object input box 611 and a query result display area 612. In addition, the medical knowledge graph query interface also includes a query control 613, a related entity drop-down menu 614, an update control 615, a download control 616, and an N-hop query control 617.

[0057] S502: When a medical entity to be queried is detected in the query object input box, check whether the medical entity to be queried is incorrect.

[0058] If the medical entity to be queried contains an error, proceed to step S503; if the medical entity to be queried does not contain an error, proceed to step S504. The medical entity to be queried in the query object input box can be entered by the user or selected by the user; that is, the query object input box is a drop-down menu, and the user can click the drop-down selection box to select the medical entity they want to query. For example, inputting medical entities such as "diabetes" or "sulbactam".

[0059] In one possible implementation, the conditions for triggering the execution of the error detection for the queried medical entity include the following two: 1. Execution is triggered when the user enters text information in the query object input box; that is, as long as text information is detected in the query object input box, the text information is checked for correctness. 2. Execution is triggered when the user enters text information in the query object input box and clicks the query control (e.g., ...). Figure 6 The query control 613 in the middle will trigger the execution.

[0060] In one possible implementation, the specific process of detecting whether the medical entity to be queried is incorrect is to compare the entity to be queried with the medical entities stored in the medical information system. This can also be understood as traversing the medical entity to be queried among the medical entities stored in the medical information system. If the medical entity to be queried can be found in the medical information system, it means that the medical entity to be queried entered in the query object input box is correct. If the medical entity to be queried cannot be found in the medical information system, it means that the medical entity to be queried entered in the query object input box is incorrect.

[0061] For example, when the query object and the input box contain "sulbactam", the medical entity is found to be incorrectly represented after testing. The standard representation should be "sulbactam sodium".

[0062] S503: Determine M related medical entities from the recorded medical entities that are associated with the medical entity to be queried, and determine the target medical entity among the M related medical entities that has the highest degree of association with the medical entity to be queried, and update the medical entity to be queried to the target medical entity.

[0063] In one possible implementation, when an error is detected in the user's input of the medical entity to be queried in the query object input box, M related medical entities are identified from the recorded medical entities and displayed in the related medical entity display area. This display could be in the form of a drop-down list, which the user can click to see. The M related medical entities are ordered, with those most closely related to the medical entity to be queried appearing first and having higher priority. Essentially, when the user enters an error in the query object input box, the related medical entity with the highest correlation is selected as the correct representation of the medical entity to be checked; that is, the related medical entity with the highest correlation is selected as the target medical entity, and the medical entity to be queried is updated to the target medical entity.

[0064] In one possible implementation, the process of determining M associated medical entities from the recorded medical entities that are related to the medical entity to be queried may include: constructing a first index list of the medical entities to be queried, the first index list including multiple index terms, each index term consisting of adjacent characters contained in the medical entity to be queried; comparing the index terms in the first index list with the index terms in the second index lists of the recorded medical entities to determine the number of co-occurrences of the index terms between the medical entity to be queried and each of the second index lists; determining multiple candidate medical entities from the recorded medical entities based on the number of co-occurrences, and determining the matching degree between the medical entity to be queried and each candidate medical entity; determining M candidate medical entities from the multiple candidate medical entities based on the matching degree, and identifying the M candidate medical entities as the M associated medical entities related to the medical entity to be queried.

[0065] For example, if the medical entity to be queried is "headache", the corresponding first index list for "headache" is constructed as: <head><swelling><pain>. Then, the second index list of all medical entities included in the medical knowledge graph is obtained, with one second index list corresponding to one medical entity. For example, "headache" is one of the entities recorded in the medical knowledge graph, and its corresponding second index list is: <head><ache><pain>. Then, the first index list (<head><swelling><pain>) and the second index list (<head><ache><pain>) are compared, and the co-occurrence frequency of the two is found to be 1, i.e., <head>. Therefore, "headache" can be considered as one of the candidate medical entities for "headache". As another example, for the medical entity "right-sided headache", the corresponding second index list is: <head><right><right side><swelling><pain>, and the co-occurrence frequency of "headache" and "right-sided headache" is found to be 2, i.e., <head><pain>. Assuming there are N medical entities recorded in the medical knowledge graph, following the methods described above, determine the co-occurrence frequency of "headache" with these N medical entities and sort them in reverse order, with the entity with the fewest co-occurrences at the end. Then, as needed, select the top I entities from the N medical entities as candidate medical entities. Calculate the matching degree (or similarity) between "headache" and each of these I candidate medical entities. Entities with a matching degree greater than a set threshold are considered related medical entities. Based on this method, M related medical entities are determined from the I candidate medical entities.

[0066] like Figure 6 As shown in 620, the user enters "sulbactam" in the query object input box 621. After detection, it is found that "sulbactam" is an incorrect expression. At this time, multiple related medical entities are displayed in the related entity drop-down menu 624. Among them, "sulbactam sodium" can be seen as the first one. Therefore, when the user clicks the query control 623, the medical knowledge graph of the medical entity "sulbactam sodium" will be displayed in the query result display area 622.

[0067] Furthermore, based on the M associated medical entities related to the medical entity to be queried, clicking on any one of these associated medical entities will display the medical knowledge graph of that associated medical entity in the query results display area. This can be understood as the medical knowledge graph query interface also including an associated entity drop-down menu and an update control associated with it. Once the M associated medical entities are identified, they are displayed in the associated entity drop-down menu. When the update control is triggered, the current medical knowledge graph displayed in the query results display area is updated to match the selected medical entity. The selected medical entity refers to the associated medical entity chosen from the M associated medical entities.

[0068] like Figure 7 As shown in 710, when the user enters "Sulbactam" in the query object input box 711, since "Sulbactam" is an incorrect expression, the query results display area shows the medical knowledge graph of the correct expression "Sulbactam Sodium". Simultaneously, the associated entity drop-down menu 713 displays related medical entities associated with "Sulbactam", including "Sulbactam Sodium", "Sulbactam", and "Sulbactam Sodium for Injection". Then, when the user selects "Sulbactam" from the associated entity drop-down menu 713 and clicks the update control 714, the medical knowledge graph in the query results display area 712 is updated. The update page is shown below. Figure 7 As shown in 720, the medical knowledge graph of "sulbactam" is displayed in the query results display area 722.

[0069] S504: Display matching medical knowledge graphs that match the medical entity being queried in the query results display area.

[0070] In one possible implementation, when an error exists in the medical entity being queried, a matching medical knowledge graph that matches the queried medical entity is displayed in the query results area. This replaces the queried medical entity with the target medical entity, and the displayed data is the target medical entity's medical knowledge graph. Figure 6 As shown, if the medical entity "Sulbactam" is an incorrect representation, then the medical knowledge graph of the medical entity "Sulbactam Sodium" will be displayed.

[0071] In one possible implementation, when no errors are found in the medical entity being queried, the matching medical knowledge graph data that matches the queried medical entity is displayed in the query results display area, such as... Figure 3 As shown, the medical entity "diabetes" is correct, so what is displayed is the medical knowledge graph of "diabetes".

[0072] Furthermore, the medical knowledge graph query interface also includes an N-hop query control, where N is a positive integer greater than 1, such as... Figure 6 The example 617: "2-hop" indicates a 2-hop query control. When the N-hop query control is not triggered, the first matching medical knowledge graph that matches the medical entity object to be queried is displayed in the query results display area. The first matching medical knowledge graph is centered on the medical entity object to be queried and includes medical entities that have a direct relationship with the medical entity object to be queried. The N-hop query control not being triggered can be understood as the default value, i.e., 1 hop. The first matching medical knowledge graph can be understood as... Figure 3 The medical knowledge graph shown is centered on "diabetes," with a ring of medical entities directly related to "diabetes" surrounding it.

[0073] When the N-hop query control is triggered, a second matching medical knowledge graph that matches the medical entity object being queried is displayed in the query results area. This second matching medical knowledge graph is centered on the medical entity object being queried and includes medical entities that are directly related to it, as well as those that are indirectly related. It can be understood as centered on "diabetes," connecting two medical entity nodes outwards. For example, "diabetes" is directly connected to "Xiaotangling Granules," and then another medical entity is connected directly to "Xiaotangling Granules."

[0074] In one possible implementation, each medical knowledge graph has a central point, which can be the medical entity to be queried. The central point's display format differs from that of other medical entities in the knowledge graph. The queried medical entity can be highlighted or represented by a color distinct from other medical entities; for example, the queried medical entity could be represented in pink, while other medical entities could be represented in black. Furthermore, the display format of other medical entities (nodes) connected to the queried medical entity can also differ. Different types of medical entities can be distinguished using different colors and shapes. Medical entities of the same category are represented by the same color and the same shape. For example... Figure 3 As shown, with "diabetes" as the center, "diabetes" is highlighted, and the surrounding nodes and medical entities of the same type are represented by the same color. For example, drugs for "diabetes", including "Xiaotangling Granules", "HbA1c Test Reagent", "Dexamethasone Acetate Cream", and "Xiaotangling Capsules", are all displayed in red; similarly, tests for "diabetes", including "Blood Glucose", "Hemorheology Test", "Serum Protein Non-enzyme", and "Serum C-peptide", are all displayed in blue; others are not listed individually, but those of the same type are displayed in the same color.

[0075] In one possible implementation, if any medical entity in the matching medical knowledge graph that matches the medical entity to be queried is selected, the attribute information of that medical entity can be displayed in the attribute information display area of ​​the medical knowledge graph query interface. For each medical entity, there is a corresponding attribute information display area. When the user selects a medical entity, its attribute information will be displayed in the attribute information display area, which is generally adjacent to the corresponding medical entity. For example... Figure 7 As shown in 721 of 720, when the user clicks on "bronchitis", the attribute information of "bronchitis" is displayed next to "bronchitis", including: the physical anatomical system is the respiratory system and the physical anatomical organ is the respiratory tract.

[0076] Optionally, the medical knowledge graph query interface may also include a download control, such as... Figure 6 The download control 616 shown allows users to download the currently visualized medical knowledge graph for convenient secondary use.

[0077] This application primarily describes a medical knowledge graph query interface. Through this interface, users can input various medical entities to query, and can also select and update the medical knowledge graph based on provided related medical entities. Furthermore, different display methods can be used to intuitively show the differences between various medical entities in the matching medical knowledge graph that match the queried medical entity, allowing users to gain a timely and comprehensive understanding of the medical knowledge graph.

[0078] Please see Figure 8 This is a flowchart illustrating another medical information display method disclosed in an embodiment of this application. Figure 8 The method for displaying medical information shown is illustrated using a medical information query interface as a medical ontology graph query interface, and includes, but is not limited to, the following steps:

[0079] S801: Displays the medical ontology graph query interface, which includes a query object input box and a query result display area.

[0080] like Figure 9 As shown in 910, when the user clicks "Ontology Atlas," a medical ontology atlas query interface will be displayed, including a query object input box 911 and a query result display area 912. In addition, the medical ontology atlas query interface also includes a query control 913 and a download control 914.

[0081] S802: Display the matching medical ontology graph that matches the medical concept object to be queried in the query results display area.

[0082] In one possible implementation, when a medical concept object to be queried exists in the query input box, a matching medical ontology graph that matches the queried medical concept object is displayed in the query results display area. This process can either involve directly displaying the matching medical ontology graph when a medical concept object is detected in the query input box, or displaying the matching medical ontology graph when a medical concept object is detected in the query input box and the corresponding query control is selected.

[0083] like Figure 9As shown in 920, when the user enters the medical concept object "e_doctor" in the query object input box 921 and clicks the query control item 923, the corresponding medical ontology map matching "e_doctor" can be seen in the query result display area 922.

[0084] In one possible implementation, the query results display area can display matching medical ontology graphs that correspond to the queried medical concept object, using a set display method. This set display method can include one or more of the following: one is that the size of the node icons in the medical ontology graph is related to the number of medical entities contained in the medical concept object corresponding to the node icon in the current medical knowledge graph. This display method is visually represented in the interface as follows: the size of the nodes corresponding to different medical concept objects is not consistent; the node size indicates how many medical entities the medical concept object covers in the current knowledge graph, i.e., the broader the range of medical entities, the larger the node. For example... Figure 10 As shown, the node sizes of "e_device" displayed in 1001 and "e_examination" displayed in 1002 are different, indicating that "e_examination" covers more medical entities in the medical knowledge graph.

[0085] Another type is the connection feature of the line linking the two node icons in the medical ontology graph. This feature is related to the frequency of the meta-relationship between the two medical concept objects corresponding to the two node icons in the current knowledge graph. Connection features include one or more of the following: color, length, thickness, and line type. For example, the darker the color, the higher the frequency of that meta-relationship in the medical knowledge graph. Figure 10 As shown in 1003, the colors of "doctor", "specializes in diseases", and "disease" are darker, indicating that there are more corresponding triples for this relationship triple in the medical knowledge graph.

[0086] In one possible implementation, when the meta-relationship between two medical concept objects is too complex, different meta-relationships can be visualized with spacing in the query results display area to achieve the purpose of intuitive display. Furthermore, due to the different colors of the relationship edges, it is easy to see which relationships between the two medical concept objects are relatively important (which can be understood as high frequency of use) and which relationships are relatively unimportant (which can be understood as low frequency of use).

[0087] like Figure 11As shown, the diagram, centered on "e_drug," displays the medical ontology of "e_drug." This diagram reveals a complex set of meta-relationships between "e_disease" and "e_drug," with varying edge lengths for different meta-relationships, facilitating a clear visual representation. Furthermore, the edge colors also differ, with the edge corresponding to the meta-relationship "e_disease," "r_contraindications," and "e_drug" being darker, indicating its greater importance and frequent use in medical knowledge graphs.

[0088] In one possible implementation, if any medical concept object in the matching medical ontology graph that matches the medical concept object to be queried is selected, the attribute information of that medical concept object can be displayed in the attribute information display area of ​​the medical ontology graph query interface. For each medical concept object, there is a corresponding attribute information display area. When the user selects a medical concept object, its attribute information will be displayed in the attribute information display area, which is generally adjacent to the corresponding medical concept object. For example... Figure 4 As shown in 403, when a user clicks "e_symptom", the attribute information of "e_symptom" is displayed next to "e_symptom", including: clinical manifestations, common causes, symptom and sign overview and many other attributes. When there is too much attribute information, the display area of ​​this attribute information may not be fully displayed, and the undisplayed content is hidden. When the user clicks again, it can be displayed in another way, such as opening another interface, or popping up a new interface on the far right of the medical ontology query interface to display the complete attribute information.

[0089] Optionally, the medical ontology atlas query interface may also include a download control, such as... Figure 9 The download control 914 shown allows users to download the currently visualized medical ontology map for convenient secondary use.

[0090] This application primarily describes a medical ontology graph query interface. Through this interface, users can input the medical concept object to be queried to search for the medical ontology. Furthermore, various design details within the interface (such as the size of the nodes corresponding to the medical concept object and the color intensity of the meta-relation edges) help users quickly understand the current medical ontology and rapidly locate high-frequency, important concepts and meta-relationships. This rapid location is crucial for the construction and subsequent application of medical knowledge graphs.

[0091] In another possible implementation, the medical information system, in addition to the medical information query interface (medical knowledge graph query interface and medical ontology graph query interface), also includes other pages, such as a medical information statistics page and a user feedback page. Simply put, this medical information system can be used... Figure 12 The explanation includes the navigation bar in section 1201, which displays the names of various interfaces, such as "Knowledge Graph," "Ontology Graph," "Question and Answer," "Statistics," and "User Feedback." When a user selects any name in the navigation bar, the corresponding interface will be displayed. Additionally, each interface in the medical information system includes links to related tools, for example... Figure 12 As shown in 1202, "IWIKI-PAGE" is the address of the medical information system's documentation, "GIT-REPO" is the address of the medical information system's code repository, and "ABOUT-US" is the address of more informational content. Each interface of the medical information system also includes displays and reminders of key events, such as... Figure 12 As shown in 1203, clicking this option will display that the current version of the medical knowledge graph is v1.2.6 and the version of the medical information system is v1.4.

[0092] In one possible implementation, when the user clicks "Statistics" in the navigation bar, a statistics interface is displayed, showing the version information of the medical knowledge graph, as shown in Figure 13. This is a diagram of a medical information statistics interface disclosed in an embodiment of this application. Figure 13 As shown, the statistics interface can display medical knowledge graph statistics, mainly showing the graph version, number of entities, number of relation triples, etc.; it can also display medical information statistics in the form of charts, showing the number of medical entity types and the number of medical entities under each version; it can also show the number of relation types and the number of ternary relations under each version, etc. Figure 13 The interface does not display all the information, only a portion of the statistical information. It's important to note that the statistical results shown are for illustrative purposes and do not fully represent the actual knowledge graph. The medical information statistics interface primarily demonstrates the iteration process between different versions of the medical knowledge graph, facilitating user understanding and management. It also allows users to add annotations after a specific version for easier management.

[0093] In one possible implementation, when a user clicks "User Feedback," a user feedback information page will be displayed. This page is mainly used to collect user feedback on the use of this medical information system.

[0094] The above describes the front-end interface (mainly the visual interface). In addition, this application also involves back-end technology, the back-end technology framework as follows: Figure 14 As shown, it includes the front-end interface and API (antv / g6, Vue.js, node.js), the back-end algorithm engine (Sanic) in the middle, and the database on the right.

[0095] 1. Front-end Interface and Interface: The most important part of the front-end interface is the formal representation of medical knowledge in the display method. This application uses the open-source visualization engine antv / g6, which has powerful visualization capabilities for network structure data. Its open-source nature avoids copyright risks, and its pure front-end nature allows for full decoupling from back-end computation, achieving flexible adaptation and deployment. The code snippet shown below is an example of the antv / g6 visualization engine:

[0096] this.graph = new G6.Graph({

[0097] container:'kgCanvas',

[0098] width:width,

[0099] height:height,

[0100] animate:false,

[0101] fitView:true,

[0102] plugins:[tooltip],

[0103] layout:{

[0104] type:'gForce',

[0105] linkDistance:250,

[0106] },

[0107] modes:{

[0108] default:['zoom-canvas','drag-canvas','drag-node'],

[0109] },

[0110] defaultNode:{

[0111] size:[75,75],

[0112] },

[0113] defaultEdge:{

[0114] type:'quadratic',

[0115] },

[0116] }]

[0117] Besides the rendering methods mentioned above, rendering can also be performed using the open-source front-end framework Vue.js. After the front-end interface development is complete, this application uses Vue-router to map the API interface to the address of the back-end algorithm engine. For example, for the data reading function, the front-end interface accesses the back-end address (port 5000) through the API, thus enabling effective data communication between the front-end and back-end. After the entire front-end interface development is complete, we can deploy the front-end interface on a machine using the lightweight web server nginx and assign it a specific domain name. This allows users / clients to access the medical knowledge management system through the specified domain name.

[0118] 2. Backend Algorithm Engine: This includes three parts: graph data reading, graph indexing algorithm, and entity linking algorithm. Graph (medical knowledge graph and ontology knowledge graph) data reading is related to the specific database selection. The graph indexing algorithm focuses more on constructing a suitable data structure that is well-compatible with the characteristics of web data, achieving the goal of fast and convenient querying. Therefore, this application focuses on the entity linking algorithm.

[0119] Entity linking associates an entity with its corresponding entity in a given medical knowledge graph. The mention identified by the entity recognition module in the given medical knowledge graph is mapped onto the knowledge graph using linking techniques to obtain richer knowledge structure information and to locate the range of candidate entities. This can correspond to the M associated medical entities identified above. For example, entity linking can link the user-inputted "sulbactam" to the entity "sulbactam sodium". The entity linking module includes two steps: medical candidate entity generation and medical entity matching. This process can be performed by a computer device or a server connected to the computer device.

[0120] 1) Medical Candidate Entity Generation: Due to the large number of entities in the knowledge graph, calculating the similarity between each input mention (which can be understood as the medical entity to be queried) and each medical entity in the medical knowledge graph before selecting the target entity would result in very high computational costs, especially given the large number of entities in the medical knowledge graph. Therefore, this application proposes a recall module based on inverted index and co-occurrence calculation to recall candidate entities, reducing the computational cost of the next step, namely similarity calculation. The specific architecture diagram of the recall module is shown below. Figure 15 As shown: At the database level, an inverted index list is first built for all medical entity terms. This application only constructs binary indexes (i.e., indexes with a string length of 2). For example, the binary index list for "headache" is: <headache>, and the binary index list for "head pain" is: <head><pain><pain>. Besides constructing index terms with a string length of 2, index terms with a string length of 3 can also be constructed, depending on the specific situation. This inverted index calculation can be performed during the construction of the medical knowledge graph; therefore, this computation can be completed offline and stored beforehand, thus avoiding introducing additional computation time during actual use.

[0121] At the input mention level, the same inverted index calculation is performed. For example, if the input mention is "headache", the corresponding binary index list is: <head><headache><pain>. After obtaining the inverted index lists of all medical entities and the input mention, the number of times the same index phrases exist between the two index lists is counted. This is called the co-occurrence count, which is how many binary index phrases co-occur between the inverted index list of a certain medical entity and the inverted index list of the input mention. For example, "headache" and "headache" have 1 co-occurrence index phrase <head>, which means the co-occurrence count is 1. After inverting these co-occurrence counts, a candidate medical entity list is obtained. Then, a preset number of medical entities are selected as needed (it can be 100, 50, or even fewer candidate medical entities as needed) for medical entity matching calculation. It should be noted that since this module only requires online calculation for the inverted index calculation and co-occurrence count of the input mention, this application can better handle the efficiency problem of entity linking in large-scale medical knowledge graphs.

[0122] 2) Medical Entity Matching: In this process, a metric can be used to calculate the matching degree between the input mention and each medical candidate entity, selecting the entity with the highest matching degree as the entity pointed to by the input mention. The metric can be cosine similarity. When determining the matching degree, this application uses a semantic matching dimension, employing a language model to compensate for the significant semantic differences caused by subtle literal variations. A pre-trained language model, MedBERT, can be used to measure the semantic relevance between the input mention and the medical entity. Generally, pre-training can obtain different feature representations based on the text context, and it has been sufficiently pre-trained on a large-scale corpus, achieving better results compared to word vectors.

[0123] Please see Figure 16 This is a model architecture diagram of a medical entity matching stage disclosed in an embodiment of this application. Figure 16 As shown in Figure 1610, this represents the model training phase, where the input mention and medical entities are fed into the language model to obtain semantic features of equal length. In this application, contrastive learning is introduced to amplify the differences between medical entities with different meanings. The specific training loss function is given by formula (1):

[0124] (1)

[0125] Where i and j are positive samples in a batch during training, i is the link result of the medical entity corresponding to i and j is the input mention corresponding to j; N is the length of the batch (usually 128); k is all other entities in the batch except for the medical entity i. These are hyperparameters, and their function is to adjust the degree to which the model pays attention to negative samples. The smaller the value (typically 0.1), the more the model focuses on negative samples that are most similar to the current sample. The model's training data consists of manually labeled entity links, including all medical entities and mappings from input mentions to entities.

[0126] After training, the prediction phase, as shown in Figure 1620, can be executed. From the training phase, we know that the model can obtain the semantic features of the input mention and the semantic features of candidate medical entities. Then, the cosine similarity between these two semantic features is calculated to determine which medical entity is semantically most similar to the input mention; that is, the greater the similarity, the closer the semantics. In this process, since the input of this phase is not all medical entities but candidate medical entities, the time cost of this phase can handle the medical entity links in a large-scale medical knowledge graph.

[0127] After the three algorithms on the engine side are developed, this application runs the above algorithms on a backend address (e.g., port 5000) through the web server provided by Sanic, so that the frontend can read and write data by accessing port 5000.

[0128] In one alternative embodiment, the backend engine algorithm can also be horizontally scaled to more AI algorithms, such as the KBQA knowledge-based question-answering algorithm, and integrated with them through the same front-end and back-end data communication methods.

[0129] 3. Database: Due to its excellent architecture design and decoupling between the front end and the back end, the database selection of this medical information system is relatively flexible. It is compatible with the following data storage methods: offline triple files, relational databases such as MySQL, and graph databases.

[0130] Using a graph database as an example, let's illustrate how to perform subgraph queries and indexes in a medical knowledge graph. For instance, if the user inputs "sulbactam," and the linked entity in the medical knowledge graph is "sulbactam sodium," the query engine will automatically generate a Cypher query statement:

[0131] MATCH(v1)-[e1]-(v2)WHERE id(v1)=='xxxxx'RETURN e1,v2 LIMIT 50

[0132] This means querying a subgraph of the medical knowledge graph that has one hop around "sulbactam sodium" (where xxxxx is the UUID of the medical entity sulbactam sodium). Alternatively, you can query a two-hop subgraph based on a user-specified number of hops.

[0133] MATCH(v1)- [e1]- (v2)-[e2]- (v3)WHEREid(v1)

[0134] =='xxxxx'RETURN e1,v2,e2,v3LIMIT 50

[0135] This means querying the subgraph around "Sulbactam Sodium" in the medical knowledge graph, which has 2 hops (where xxxxx is the UUID of the medical entity Sulbactam Sodium).

[0136] Notably, Cypher, as a widely used graph database query language, is supported by most graph databases, and its excellent compatibility greatly expands the applicability of this invention. Furthermore, since different projects have different delivery requirements and hardware conditions, this application can also support other database selections, which is crucial for project delivery. For example, for projects like the Shenzhen CDC, this application can use a MySQL relational database to adapt to the client's existing database structure as much as possible. However, for self-developed businesses like the bioinformatics cloud, a triplet file format can be used to adapt to rapid data iteration as much as possible.

[0137] This application embodiment mainly describes the backend technology. As can be seen from the above description, the medical information system is entirely based on open-source technical components and self-developed algorithms, which can guarantee the copyright and security of the product, get rid of the copyright risks brought about by certain commercial technical components, and due to its excellent architecture design and complete decoupling of the front end and back end, the system has the ability to be flexibly deployed and flexibly expanded to meet the specific needs of different customers.

[0138] Based on the above method embodiments, this application also provides a schematic diagram of the structure of a medical information display device. See also... Figure 17 This is a schematic diagram of the structure of a medical information display device provided in an embodiment of this application. Figure 17 The medical information display device 1700 shown can operate the following units:

[0139] Display unit 1701 is used to display a medical information query interface, which includes a query object input box and a query result display area;

[0140] The display unit 1701 is also used to display medical information query results that match the target query object in the query result display area, wherein the target query object is the query object entered in the query object input box;

[0141] The medical information query interface can be either a medical knowledge graph query interface or a medical ontology graph query interface. When the medical information query interface is a medical knowledge graph query interface, the target query object is the medical entity to be queried, and the medical information query result is a matching medical knowledge graph that matches the medical entity to be queried. When the medical information query interface is a medical ontology graph query interface, the target query object is the medical concept object to be queried, and the medical information query result is a matching medical ontology graph that matches the medical concept object to be queried.

[0142] In one possible implementation, the medical information query interface is the medical knowledge graph query interface, and when the display unit 1701 displays the medical information query results matching the target query object in the query result display area, it is specifically used for:

[0143] When an error is detected in the medical entity to be queried, M associated medical entities are determined from the recorded medical entities that are related to the medical entity to be queried, where M is a positive integer;

[0144] Identify the target medical entity among the M associated medical entities that has the highest correlation with the medical entity to be queried;

[0145] The query results display area displays a matching medical knowledge graph that matches the target medical entity.

[0146] In one possible implementation, the medical information query interface is the medical knowledge graph query interface, which further includes an N-hop query control, where N is a positive integer greater than 1; when the display unit 1701 displays the medical information query results matching the target query object in the query result display area, it is specifically used for:

[0147] When the N-hop query control is not triggered, a first matching medical knowledge graph that matches the medical entity object to be queried is displayed in the query result display area; the first matching medical knowledge graph is centered on the medical entity object to be queried and includes medical entities that have a direct relationship with the medical entity object to be queried.

[0148] When the N-hop query control is triggered, a second matching medical knowledge graph that matches the medical entity object to be queried is displayed in the query result display area; the second matching medical knowledge graph is centered on the medical entity object to be queried, and includes medical entities that have a direct relationship with the medical entity object to be queried, as well as medical entities that have an indirect relationship with the medical entity object to be queried.

[0149] In one possible implementation, the medical knowledge graph query interface further includes a drop-down menu of associated entities and an update control associated with the drop-down menu of associated entities, and the display unit 1701 is further configured to:

[0150] The M associated medical entities are displayed in the associated entity drop-down menu;

[0151] When the update control is triggered, the current medical knowledge graph displayed in the query results display area will be updated to a matching medical knowledge graph that matches the selected medical entity; the selected medical entity is the associated medical entity selected in the selection operation of the M associated medical entities.

[0152] In one possible implementation, the medical information query interface is a medical ontology graph query interface. When the display unit 1701 displays medical information query results matching the target query object in the query result display area, it is specifically used for:

[0153] The matching medical ontology map that matches the medical concept object to be queried is displayed in the query results display area in a set display mode;

[0154] The display settings include one or more of the following: the size of the node icon in the medical ontology graph is related to the number of medical entities contained in the medical concept object corresponding to the node icon in the current medical knowledge graph; the connection feature of the line connecting the relationship between two node icons in the medical ontology graph is related to the frequency of the meta-relationship between the two medical concept objects corresponding to the two node icons in the current knowledge graph; the connection feature includes one or more of the following: color, length, thickness, and line type.

[0155] In one possible implementation, the display unit 1701 is further configured to:

[0156] When the medical information query result is a matching medical knowledge graph that matches the medical entity to be queried, if any medical entity in the matching medical knowledge graph is selected, the attribute information of the medical entity is displayed in the attribute information display area of ​​the medical knowledge graph query interface.

[0157] When the medical information query result is a matching medical ontology graph that matches the medical concept object to be queried, if any medical concept object in the matching medical ontology graph is selected, the attribute information of any medical concept object will be displayed in the attribute information display area of ​​the medical ontology graph query interface.

[0158] In one possible implementation, when determining M associated medical entities related to the medical entity to be queried from the recorded medical entities, the determining unit 1702 is specifically used for:

[0159] Construct a first index list of the medical entities to be queried, the first index list including multiple index terms, each index term consisting of adjacent characters contained in the medical entity to be queried;

[0160] The index terms in the first index list are compared with the index terms in the second index list of each recorded medical entity to determine the number of times the index terms of the medical entity to be queried co-occur with each of the second index tables.

[0161] Based on the co-occurrence count, multiple candidate medical entities are determined from the recorded medical entities, and the matching degree between the medical entity to be queried and each of the candidate medical entities is determined.

[0162] Based on the matching degree, M candidate medical entities are determined from the plurality of candidate medical entities, and the M candidate medical entities are determined as M associated medical entities related to the medical entity to be queried.

[0163] It is understood that the functions of each functional unit of the medical information display device provided in the embodiments of this application can be specifically implemented according to the methods in the above method embodiments, and the specific implementation process can be referred to the relevant descriptions in the above method embodiments, which will not be repeated here.

[0164] In feasible embodiments, the medical information display device provided in this application can be implemented in software. The medical information display device can be stored in a memory and can be software in the form of programs and plug-ins, and includes a series of units, including a display unit and a determination unit; wherein, the display unit and the determination unit are used to implement the medical information display method provided in this application.

[0165] In other feasible embodiments, the medical information display device provided in this application embodiment can also be implemented in a combination of hardware and software. As an example, the medical information display device provided in this application embodiment can be a processor in the form of a hardware decoding processor, which is programmed to execute the medical information display method provided in this application embodiment. For example, the processor in the form of a hardware decoding processor can be one or more application-specific integrated circuits (ASICs), DSPs, programmable logic devices (PLDs), complex programmable logic devices (CPLDs), field-programmable gate arrays (FPGAs), or other electronic components.

[0166] The embodiments of this application display a medical information query interface, which includes a query object input box and a query result display box. When a target query object is entered in the query object input box, the medical information query results matching the target query object are displayed in the query result display area. The medical information query interface can be a medical knowledge graph query interface or a medical ontology graph query interface. When the medical information query interface is a medical knowledge graph query interface, the target query object is the medical entity to be queried, and the medical information query results are matching medical knowledge graphs that match the medical entity to be queried. Based on this medical knowledge graph query interface, various medical entities can be queried, enabling medical workers to quickly grasp the medical knowledge graphs corresponding to each medical entity. When the medical information query interface is a medical ontology graph query interface, the target query object is the medical concept object to be queried, and the medical information query results are matching medical ontology graphs that match the medical concept object to be queried. Based on this medical ontology graph query interface, various medical concept objects can be queried, helping medical workers quickly understand the current medical ontology and quickly locate high-frequency important medical concepts and meta-relationships. Therefore, the medical information display method proposed in this application can realize the query of medical information and enrich the types of medical information available for query.

[0167] Please see Figure 18 , Figure 18 This is a schematic diagram of a computer device provided in an embodiment of this application. The computer device described in this embodiment includes a processor 1801, a communication interface 1802, and a memory 1803. The processor 1801, communication interface 1802, and memory 1803 can be connected via a bus or other means; this embodiment uses a bus connection as an example.

[0168] The processor 1801 (or CPU, Central Processing Unit) is the computing and control core of the computer device. It can parse various instructions and process various data within the computer device. For example, the CPU can parse power-on / off commands sent by the user and control the computer device to perform power-on / off operations; it can also transmit various interactive data between internal structures of the computer device. The communication interface 1802 may optionally include standard wired interfaces or wireless interfaces (such as Wi-Fi, mobile communication interfaces, etc.), and is controlled by the processor 1801 for sending and receiving data. The memory 1803 is the storage device in the computer device, used to store programs and data. It is understood that the memory 1803 here can include the computer device's built-in memory, or it can include extended memory supported by the computer device. The memory 1803 provides storage space for the computer device's operating system, which may include, but is not limited to, Android, iOS, Windows Phone, etc., and this application does not limit this.

[0169] In this embodiment of the application, the processor 1801 performs the following operations by running the executable program code in the memory 1803:

[0170] The medical information query interface includes a query object input box and a query result display area.

[0171] The query results display area displays medical information query results that match the target query object, which is the query object entered in the query object input box;

[0172] The medical information query interface can be either a medical knowledge graph query interface or a medical ontology graph query interface. When the medical information query interface is a medical knowledge graph query interface, the target query object is the medical entity to be queried, and the medical information query result is a matching medical knowledge graph that matches the medical entity to be queried. When the medical information query interface is a medical ontology graph query interface, the target query object is the medical concept object to be queried, and the medical information query result is a matching medical ontology graph that matches the medical concept object to be queried.

[0173] In one possible implementation, the medical information query interface is the medical knowledge graph query interface, and when the processor 1801 displays the medical information query results matching the target query object in the query result display area, it is specifically used for:

[0174] When an error is detected in the medical entity to be queried, M associated medical entities are determined from the recorded medical entities that are related to the medical entity to be queried, where M is a positive integer;

[0175] Identify the target medical entity among the M associated medical entities that has the highest correlation with the medical entity to be queried;

[0176] The query results display area displays a matching medical knowledge graph that matches the target medical entity.

[0177] In one possible implementation, the medical information query interface is the medical knowledge graph query interface, which further includes an N-hop query control, where N is a positive integer greater than 1; when the processor 1801 displays medical information query results matching the target query object in the query result display area, it is specifically used for:

[0178] Displaying medical information query results that match the target query object in the query results display area includes:

[0179] When the N-hop query control is not triggered, a first matching medical knowledge graph that matches the medical entity object to be queried is displayed in the query result display area; the first matching medical knowledge graph is centered on the medical entity object to be queried and includes medical entities that have a direct relationship with the medical entity object to be queried.

[0180] When the N-hop query control is triggered, a second matching medical knowledge graph that matches the medical entity object to be queried is displayed in the query result display area; the second matching medical knowledge graph is centered on the medical entity object to be queried, and includes medical entities that have a direct relationship with the medical entity object to be queried, as well as medical entities that have an indirect relationship with the medical entity object to be queried.

[0181] In one possible implementation, the medical knowledge graph query interface further includes an associated entity drop-down menu and an update control associated with the associated entity drop-down menu, and the processor 1801 is further configured to:

[0182] The M associated medical entities are displayed in the associated entity drop-down menu;

[0183] When the update control is triggered, the current medical knowledge graph displayed in the query results display area will be updated to a matching medical knowledge graph that matches the selected medical entity; the selected medical entity is the associated medical entity selected in the selection operation of the M associated medical entities.

[0184] In one possible implementation, the medical information query interface is a medical ontology graph query interface. When the processor 1801 displays medical information query results matching the target query object in the query result display area, it is specifically used for:

[0185] The matching medical ontology map that matches the medical concept object to be queried is displayed in the query results display area in a set display mode;

[0186] The display settings include one or more of the following: the size of the node icon in the medical ontology graph is related to the number of medical entities contained in the medical concept object corresponding to the node icon in the current medical knowledge graph; the connection feature of the line connecting the relationship between two node icons in the medical ontology graph is related to the frequency of the meta-relationship between the two medical concept objects corresponding to the two node icons in the current knowledge graph; the connection feature includes one or more of the following: color, length, thickness, and line type.

[0187] In one possible implementation, the processor 1801 is further configured to:

[0188] When the medical information query result is a matching medical knowledge graph that matches the medical entity to be queried, if any medical entity in the matching medical knowledge graph is selected, the attribute information of the medical entity is displayed in the attribute information display area of ​​the medical knowledge graph query interface.

[0189] When the medical information query result is a matching medical ontology graph that matches the medical concept object to be queried, if any medical concept object in the matching medical ontology graph is selected, the attribute information of any medical concept object will be displayed in the attribute information display area of ​​the medical ontology graph query interface.

[0190] In one possible implementation, when the processor 1801 determines M associated medical entities related to the queried medical entity from the recorded medical entities, it is specifically used for:

[0191] Construct a first index list of the medical entities to be queried, the first index list including multiple index terms, each index term consisting of adjacent characters contained in the medical entity to be queried;

[0192] The index terms in the first index list are compared with the index terms in the second index list of each recorded medical entity to determine the number of times the index terms of the medical entity to be queried co-occur with each of the second index tables.

[0193] Based on the co-occurrence count, multiple candidate medical entities are determined from the recorded medical entities, and the matching degree between the medical entity to be queried and each of the candidate medical entities is determined.

[0194] Based on the matching degree, M candidate medical entities are determined from the plurality of candidate medical entities, and the M candidate medical entities are determined as M associated medical entities related to the medical entity to be queried.

[0195] According to one aspect of this application, an embodiment of this application also provides a computer product including a computer program stored in a computer-readable storage medium. A processor 1801 reads the computer program from the computer-readable storage medium and executes the computer program, causing a computer device 1800 to perform... Figure 2 , Figure 5 as well as Figure 8 Methods for displaying medical information.

[0196] It should be noted that, for the sake of simplicity, the foregoing method embodiments are all described as a series of actions. However, those skilled in the art should understand that this application is not limited to the described order of actions, as some steps may be performed in other orders or simultaneously according to this application. Furthermore, those skilled in the art should also understand that the embodiments described in the specification are preferred embodiments, and the actions and modules involved are not necessarily essential to this application.

[0197] In the several embodiments provided in this application, it should be understood that the disclosed apparatus can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative, and the division of modules described above is only a logical functional division. In actual implementation, there may be other division methods. For example, multiple modules or components may be combined or integrated into another system, or some features may be ignored or not executed.

[0198] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A method for displaying medical information, characterized in that, The method includes: The medical information query interface includes a query object input box and a query result display area. The query results display area displays medical information query results that match the target query object, which is the query object entered in the query object input box. The medical information query interface can be a medical knowledge graph query interface or a medical ontology graph query interface. When the medical information query interface is a medical knowledge graph query interface, the target query object is the medical entity to be queried, and the medical information query results are matching medical knowledge graphs that match the medical entity to be queried. When the medical information query interface is a medical ontology graph query interface, the target query object is the medical concept object to be queried, and the medical information query results are matching medical ontology graphs that match the medical concept object to be queried. Where the medical information query interface is the medical knowledge graph query interface, the step of displaying medical information query results matching the target query object in the query result display area includes: when an error is detected in the medical entity to be queried, constructing a first index list of the medical entity to be queried, the first index list including multiple index phrases, each index phrase consisting of adjacent characters contained in the medical entity to be queried; comparing the index phrases in the first index list with the index phrases in the second index lists of each recorded medical entity to determine the relationship between the medical entity to be queried and each of the second index lists. The query results are analyzed as follows: The number of co-occurrences of index terms is counted; multiple candidate medical entities are determined from the recorded medical entities based on the co-occurrence counts, and the matching degree between the queried medical entity and each of the candidate medical entities is determined; M candidate medical entities are determined from the multiple candidate medical entities based on the matching degree, and the M candidate medical entities are identified as M associated medical entities related to the queried medical entity, where M is a positive integer; the target medical entity with the highest correlation degree among the M associated medical entities is determined; and a matching medical knowledge graph matching the target medical entity is displayed in the query result display area.

2. The method according to claim 1, characterized in that, The medical information query interface is the medical knowledge graph query interface, and the medical knowledge graph query interface also includes an N-hop query control, where N is a positive integer greater than 1; Displaying medical information query results that match the target query object in the query results display area includes: When the N-hop query control is not triggered, a first matching medical knowledge graph that matches the medical entity object to be queried is displayed in the query result display area; the first matching medical knowledge graph is centered on the medical entity object to be queried and includes medical entities that have a direct relationship with the medical entity object to be queried. When the N-hop query control is triggered, a second matching medical knowledge graph that matches the medical entity object to be queried is displayed in the query result display area; the second matching medical knowledge graph is centered on the medical entity object to be queried, and includes medical entities that have a direct relationship with the medical entity object to be queried, as well as medical entities that have an indirect relationship with the medical entity object to be queried.

3. The method according to claim 1, characterized in that, The medical knowledge graph query interface also includes a drop-down menu of associated entities and an update control associated with the drop-down menu of associated entities. The method further includes: The M associated medical entities are displayed in the associated entity drop-down menu; When the update control is triggered, the current medical knowledge graph displayed in the query results display area will be updated to a matching medical knowledge graph that matches the selected medical entity; the selected medical entity is the associated medical entity selected in the selection operation of the M associated medical entities.

4. The method according to claim 1, characterized in that, The medical information query interface is a medical ontology graph query interface. The display of medical information query results matching the target query object in the query results display area includes: The matching medical ontology map that matches the medical concept object to be queried is displayed in the query results display area in a set display mode; The display settings include one or more of the following: the size of the node icon in the medical ontology graph is related to the number of medical entities contained in the medical concept object corresponding to the node icon in the current medical knowledge graph; the connection feature of the line connecting the relationship between two node icons in the medical ontology graph is related to the frequency of the meta-relationship between the two medical concept objects corresponding to the two node icons in the current knowledge graph; the connection feature includes one or more of the following: color, length, thickness, and line type.

5. The method according to claim 1, characterized in that, The method further includes: When the medical information query result is a matching medical knowledge graph that matches the medical entity to be queried, if any medical entity in the matching medical knowledge graph is selected, the attribute information of the medical entity is displayed in the attribute information display area of ​​the medical knowledge graph query interface. When the medical information query result is a matching medical ontology graph that matches the medical concept object to be queried, if any medical concept object in the matching medical ontology graph is selected, the attribute information of any medical concept object will be displayed in the attribute information display area of ​​the medical ontology graph query interface.

6. A medical information display device, characterized in that, The device includes: The display unit is used to display a medical information query interface, which includes a query object input box and a query result display area. The display unit is also used to display medical information query results that match the target query object in the query result display area, wherein the target query object is the query object entered in the query object input box; The medical information query interface can be either a medical knowledge graph query interface or a medical ontology graph query interface. When the medical information query interface is a medical knowledge graph query interface, the target query object is the medical entity to be queried, and the medical information query result is a matching medical knowledge graph that matches the medical entity to be queried. When the medical information query interface is a medical ontology graph query interface, the target query object is the medical concept object to be queried, and the medical information query result is a matching medical ontology graph that matches the medical concept object to be queried. Specifically, when the medical information query interface is the medical knowledge graph query interface, and the display unit displays medical information query results matching the target query object in the query result display area, it is used to: when an error is detected in the medical entity to be queried, construct a first index list of the medical entity to be queried, the first index list including multiple index phrases, each index phrase consisting of adjacent characters contained in the medical entity to be queried; compare the index phrases in the first index list with the index phrases in the second index lists of each recorded medical entity, and determine whether the medical entity to be queried matches the index phrases in the second index lists of each recorded medical entity. The query results are analyzed as follows: The number of co-occurrences of index terms between the index lists is counted; multiple candidate medical entities are determined from the recorded medical entities based on the co-occurrence counts, and the matching degree between the queried medical entity and each of the candidate medical entities is determined; M candidate medical entities are determined from the multiple candidate medical entities based on the matching degree, and the M candidate medical entities are identified as M associated medical entities related to the queried medical entity, where M is a positive integer; the target medical entity with the highest correlation degree among the M associated medical entities is determined; and a matching medical knowledge graph matching the target medical entity is displayed in the query result display area.

7. A computer device, characterized in that, The computer device includes: A processor, suitable for implementing one or more computer programs; and, A computer storage medium storing one or more computer programs, said one or more computer programs being adapted to be loaded by the processor and executed as described in any one of claims 1-5.

8. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores one or more computer programs, which are adapted to be loaded by a processor and executed as described in any one of claims 1-5.

9. A computer program product, characterized in that, The computer program product includes a computer program stored in a computer-readable storage medium, a processor of a computer device reading the computer program from the computer-readable storage medium, and the processor executing the computer program to cause the computer device to perform the medical information display method as described in any one of claims 1-5.

Images