A method, apparatus, device and storage medium for disease phenotype classification

Abstract

The application discloses a disease phenotype classification method, device, equipment and storage medium. A disease core treatment measure item database is generated according to the corresponding relationship between each disease and each core treatment measure item in a knowledge source. A core treatment measure item category database is established according to the core treatment measure item category corresponding to each core treatment measure item in the disease core treatment measure item database. The core treatment measure item category database is optimized according to medical record data to obtain a disease phenotype classification set. In the application, the disease phenotype is distinguished based on the core treatment measure item category corresponding to the core treatment measure item, the influence of factors irrelevant to the disease in the classification process is avoided, and the accuracy of the disease phenotype classification is improved.

Description

Technical Field

[0001] This application relates to the field of information technology, and in particular to a method, apparatus, device, and storage medium for classifying disease phenotypes. Background Technology

[0002] Disease classification generally refers to the classification of diseases for statistical or educational purposes, such as classifying diseases according to organ systems (cardiovascular diseases, respiratory diseases, etc.) or etiologies (infectious diseases, non-infectious diseases, traumatic diseases, etc.). Disease phenotypic classification refers to the further classification of patients by doctors under the same diagnosis. For example, after diagnosing pneumonia, if it is determined to be lobar pneumonia caused by Streptococcus pneumoniae infection, penicillin treatment should be given instead of other antibiotics. Accurate and effective disease phenotypic classification can help doctors make better treatment decisions.

[0003] In existing technologies, disease phenotype classification mostly involves extracting or labeling all of a patient's health characteristics and then using methods such as machine learning or knowledge graphs to classify the patient's disease phenotype. However, since most disease phenotype classifications are not determined by all health characteristics, and some indirectly related health characteristics often account for a large proportion, these indirectly related health characteristics often have high weight values ​​in machine learning or knowledge graphs, affecting the classification results and reducing the accuracy of disease phenotype classification. Summary of the Invention

[0004] In view of the above problems, this application provides a method, apparatus, device and storage medium for disease phenotype classification.

[0005] The embodiments of this application disclose the following technical solutions:

[0006] Firstly, this application provides a method for classifying disease phenotypes, including:

[0007] Based on the correspondence between each disease and each core treatment measure item in the knowledge source, a database of core treatment measures items for diseases is generated;

[0008] Based on the core treatment measure item database, a core treatment measure item category database is established for each core treatment measure item corresponding to the core treatment measure item category.

[0009] Based on medical record data and the database of core treatment measures categories, diseases with the same core treatment measure category under the same diagnosis are classified as the same disease phenotype, thus obtaining a disease phenotype classification set.

[0010] Furthermore, based on medical record data and the core treatment measure category database, diseases with the same core treatment category under the same diagnosis are classified as the same disease phenotype, thus obtaining a disease phenotype classification set, including:

[0011] Based on medical record data and the database of core treatment measures, diseases with the same core treatment category under the same diagnosis are classified as the same disease phenotype, thus obtaining the original disease phenotype classification set.

[0012] Calculate the similarity between each disease phenotype in the original disease phenotype classification set;

[0013] The disease phenotypes with similarity higher than a preset value are merged to obtain the disease phenotype classification set.

[0014] Further, calculating the similarity between various disease phenotypes in the original disease phenotype classification set includes:

[0015] Based on the core treatment item category corresponding to the disease phenotype in the original disease phenotype classification set, generate the core treatment item category vector of the disease phenotype.

[0016] The similarity between disease phenotypes is calculated based on the category vector of the core treatment measures corresponding to each disease phenotype.

[0017] Furthermore, based on medical record data and the core treatment item category database, diseases with the same core treatment item category under the same diagnosis are classified as the same disease phenotype, thus obtaining a disease phenotype classification set, including:

[0018] By statistically analyzing the core treatment measures categories corresponding to each disease in the medical record data, a database of real core treatment measures categories is obtained.

[0019] Based on the database of actual core treatment measures and the database of core treatment measures, diseases with the same core treatment measure category under the same diagnosis are classified as the same disease phenotype, thus obtaining a disease phenotype classification set.

[0020] Furthermore, the core treatment measure categories corresponding to each disease in the statistical medical record data are used to obtain a database of real core treatment measure category categories, including:

[0021] The categories of core drug-related treatment measures and core surgical procedure-related treatment measures for each disease in the statistical medical record data are analyzed.

[0022] The drug-related core treatment measure category and the surgical procedure-related core treatment measure category are merged to obtain a database of real core treatment measure category.

[0023] Furthermore, it also includes:

[0024] The names of the core treatment measures categories are standardized.

[0025] Second aspect: This application provides a device for classifying disease phenotypes, comprising:

[0026] Module generation, module creation, and module acquisition;

[0027] The generation module is used to generate a database of core treatment measures for diseases based on the correspondence between each disease and each core treatment measure item in the knowledge source.

[0028] The establishment module is used to establish a core treatment measure item category database based on the core treatment measure item category corresponding to each core treatment measure item in the disease core treatment measure item database;

[0029] The acquisition module is used to classify diseases with the same core treatment item category under the same diagnosis as the same disease phenotype based on medical record data and the core treatment item category database, thereby obtaining a disease phenotype classification set.

[0030] Further, the obtaining module includes:

[0031] Acquisition unit, calculation unit, and merging unit;

[0032] The obtaining unit is used to, based on medical record data and the core treatment measure category database, classify diseases with the same core treatment category under the same diagnosis as the same disease phenotype, and obtain the original disease phenotype classification set.

[0033] The computing unit is used to calculate the similarity between various disease phenotypes in the original disease phenotype classification set;

[0034] The merging unit is used to merge disease phenotypes with similarity higher than a preset value to obtain the disease phenotype classification set.

[0035] Furthermore, the computing unit includes:

[0036] Generation unit and computation subunit;

[0037] The generation unit is used to generate a core treatment measure category vector for the disease phenotype based on the core treatment measure category corresponding to the disease phenotype in the original disease phenotype classification set.

[0038] The computational subunit is used to calculate the similarity between disease phenotypes based on the category vector of the core treatment measures corresponding to each disease phenotype.

[0039] Further, the obtaining module includes:

[0040] Statistical unit and acquisition sub-unit;

[0041] The statistical unit is used to statistically analyze the core treatment measure categories corresponding to each disease in the medical record data to obtain a database of real core treatment measure category categories.

[0042] The obtaining subunit is used to classify diseases with the same core treatment item category under the same diagnosis as the same disease phenotype based on the real core treatment item category database and the core treatment item category database, thereby obtaining a disease phenotype classification set.

[0043] Furthermore, the statistical unit includes:

[0044] Statistical sub-units and merged sub-units;

[0045] The statistical subunit is used to statistically analyze the categories of core drug-related treatment measures and core surgical procedure-related treatment measures for each disease in the medical record data.

[0046] The merging subunit is used to merge the drug-related core treatment measure category with the surgical procedure-related core treatment measure category to obtain a database of real core treatment measure category.

[0047] Furthermore, it also includes:

[0048] The names of the core treatment measures categories are standardized.

[0049] Thirdly: This application provides an electronic device, which includes a processor and a memory:

[0050] The memory is used to store program code and transmit the program code to the processor;

[0051] The processor is configured to execute, according to instructions in the program code, the steps of the method for classifying disease phenotypes as described in the first aspect above.

[0052] Fourth aspect: This application provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps of a method for classifying disease phenotypes as described in the first aspect above.

[0053] Compared with the prior art, this application has the following beneficial effects:

[0054] In this application, a database of core treatment measures for diseases is generated based on the correspondence between various diseases and core treatment measures in the knowledge source. A database of core treatment measure category is established based on the core treatment measure category corresponding to each core treatment measure in the database. Based on case data and the core treatment measure category database, diseases with the same core treatment measure category under the same diagnosis are classified as the same disease phenotype, thus obtaining a disease phenotype classification set. This application, by classifying diseases with the same core treatment measure category under the same diagnosis as the same disease phenotype based on the core treatment measure category corresponding to the core treatment measure, avoids the influence of factors with little correlation to the disease during the classification process, thereby improving the accuracy of disease phenotype classification. Attached Figure Description

[0055] 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.

[0056] Figure 1 A flowchart illustrating a method for classifying disease phenotypes provided in this application embodiment;

[0057] Figure 2 A flowchart illustrating a method for treating a combined disease phenotype, as provided in this application embodiment;

[0058] Figure 3 A schematic diagram of the structure of a device for classifying disease phenotypes provided in an embodiment of this application;

[0059] Figure 4 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Detailed Implementation

[0060] As described above, improving the accuracy of disease phenotype classification has become a technical problem that urgently needs to be solved by those skilled in the art.

[0061] Research has revealed that most existing technologies classify disease phenotypes by extracting or labeling all of a patient's health characteristics and then using machine learning or knowledge graphs. However, since most disease phenotype classifications are not determined by all health characteristics, and some indirectly related health characteristics often account for a significant proportion, these indirectly related health characteristics tend to have higher weights in machine learning or knowledge graphs. This influences the classification results and reduces the accuracy of disease phenotype classification.

[0062] This application generates a database of core treatment measures for diseases based on the correspondence between various diseases and core treatment measures in the knowledge source. It then establishes a database of core treatment measure category based on the core treatment measure category corresponding to each core treatment measure in the database. Finally, based on case data and the core treatment measure category database, diseases with the same core treatment measure category under the same diagnosis are classified as the same disease phenotype, thus obtaining a disease phenotype classification set. This application, by classifying diseases with the same core treatment measure category under the same diagnosis as the same disease phenotype based on the core treatment measure category corresponding to the core treatment measure, avoids the influence of factors with little correlation to the disease during the classification process, improving the accuracy of disease phenotype classification. Furthermore, the disease phenotype classification set can distinguish different disease phenotypes of the same disease. Since disease phenotypes usually have a more definite correlation with treatment decisions, the disease phenotype classification set obtained by the method provided in this application can classify diseases with the same treatment decisions as one category, effectively guiding treatment decisions.

[0063] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present application, and not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present application.

[0064] See Figure 1 The figure is a flowchart of a method for classifying disease phenotypes according to an embodiment of this application. This application provides a method for classifying disease phenotypes, which can be implemented through S101 to S103.

[0065] S101: Generate a database of core treatment measures for diseases based on the correspondence between each disease and each core treatment measure item in the knowledge source.

[0066] Specifically, knowledge sources can include relevant medical textbooks, literature, clinical guidelines, and relevant national documents. The related medical and health terminology can be standardized according to a medical and health standard terminology set. This medical and health standard terminology set includes Systematic Medical Nomenclature - Clinical Terminology (SNOMED CT), the International Classification of Diseases (ICD-10), and / or the Unified Medical Language System (UMLS).

[0067] Disease phenotypes exhibit a definite correlation with treatment regimens, typically indicating similar patient presentations within the same disease. Each disease involves different management interventions (treatment measures) during treatment, which can be categorized into pharmacological and surgical interventions. Specifically, pharmacological interventions include core and non-core intervention drugs. Core intervention drugs are those associated with core intervention measures, such as aspirin used to treat myocardial infarction. Non-core intervention drugs include adjuvant drugs and symptomatic treatment drugs, such as saline solution and antipyretic analgesics. Surgical interventions similarly include core and non-core surgical procedures. A disease core intervention database is generated based on the correspondence between diseases and core intervention measures from the knowledge source. This database contains the core intervention measures corresponding to each disease. For example, core intervention measures for myocardial infarction include aspirin use and coronary stent placement.

[0068] S102: Based on the core treatment measure item category corresponding to each core treatment measure item in the core treatment measure item database, establish a core treatment measure item category database.

[0069] Specifically, each core treatment measure item corresponds to a superior category. For example, aspirin belongs to the category of antiplatelet drugs, and antiplatelet drugs are the superior category of aspirin. Therefore, the core treatment measure item category corresponding to aspirin treatment is "treatment with antiplatelet drugs." Based on the core treatment measure item category database, a core treatment measure item category database is established. Thus, based on the core treatment measure item category database, the core treatment measures corresponding to each disease, and the corresponding core treatment measure item category, can be determined.

[0070] The categories of core treatment measures corresponding to each core treatment measure item can be standardized based on relevant standard datasets. For example, relevant standard datasets may include the Anatomical Therapeutic Chemical (ATC) classification system, the National Basic Medical Insurance, Work Injury Insurance and Maternity Insurance Drug Catalog (2021), the Diagnosis-Intervention Packet (DIP) catalog, and the International Classification of Diseases Surgical and Operational Codes (ICD-9-CM3), etc.

[0071] For example, in the ATC (Advanced Treatment Protocol), drugs are divided into five levels. In this embodiment, the core treatment measure item for drugs corresponds to the fifth level category in the ATC, and the core treatment measure item category corresponds to the fourth level category in the ATC. Therefore, the core treatment measure item category corresponding to the core treatment measure item for drugs can be standardized according to the fourth level category in the ATC, that is, the fourth level category name in the ATC is used as the name of the core treatment measure item category. The core treatment measure item category corresponding to the core treatment measure item for surgical procedures is standardized according to the DIP (Digital Instructions for Public Procedures) catalog to meet the needs of different versions used by different medical institutions.

[0072] S103: Based on medical record data and the core treatment item category database, diseases with the same core treatment item category under the same diagnosis are classified as the same disease phenotype, and a disease phenotype classification set is obtained.

[0073] Based on medical record data and the database of core treatment measures categories, different phenotypes of the same disease can be distinguished. Diseases with the same core treatment category under the same diagnosis are classified as the same disease phenotype, resulting in a disease phenotype classification set. Specifically, case data can come from the actual treatment process, such as the sum of text, symbols, charts, images, slides, and other materials generated by medical staff during medical activities.

[0074] In summary, this application distinguishes disease phenotypes based on the categories of core treatment measures corresponding to core treatment measures, avoiding the influence of factors with little correlation to the disease during the classification process and improving the accuracy of disease phenotype classification. Furthermore, within the disease phenotype classification set, different disease phenotypes of the same disease can be distinguished. Since disease phenotypes usually have a more definite correlation with treatment decisions, the disease phenotype classification set obtained through the method provided in this application can classify diseases with the same treatment decisions as one category, effectively guiding treatment decisions.

[0075] Furthermore, the above-mentioned S103 can be specifically implemented through S201 to S202.

[0076] S201: Statistically analyze the core treatment measures categories corresponding to each disease in the medical record data to obtain a database of real core treatment measures categories.

[0077] S202: Based on the database of actual core treatment measures and the database of core treatment measures, diseases with the same core treatment measure category under the same diagnosis are classified as the same disease phenotype, and a disease phenotype classification set is obtained.

[0078] Specifically, medical record data contains a large amount of information, including diagnosis, symptoms, examinations and tests, past medical history, medications, and surgical procedures. However, patient characteristics may not be fully recorded in the medical record data. What is accurately and clearly recorded in the medical record information is the diagnosis and treatment measures taken for that specific case. This information can be filtered using a core treatment measure category database to obtain a database of true core treatment measure category databases. Based on this database of true core treatment measure category databases, the database is then optimized to obtain a disease phenotype classification set.

[0079] Compared with existing classification methods that use all health data features of patients as basic classification data elements, this application distinguishes disease phenotypes based on the core treatment measure category corresponding to the core treatment measure item. At the same time, it optimizes the core treatment measure category database according to the real core treatment measure category database to obtain a disease phenotype classification set, eliminating the interference of non-critical influencing factors and ensuring the accuracy and reliability of disease phenotype classification.

[0080] Furthermore, the core treatment measure categories corresponding to each disease in the statistical medical record data are used to obtain a database of real core treatment measure category categories, including:

[0081] The statistical medical record data includes the categories of core treatment measures related to medication and core treatment measures related to surgical procedures for each disease.

[0082] Specifically, since some terms in the medical record data may not be consistent with the terms in the core treatment measure category database, the terms in the medical record data can be standardized, and the drug-related core treatment measure category and surgical procedure-related core treatment measure category corresponding to each disease in the medical record data can be statistically analyzed.

[0083] The drug-related core treatment measure category and the surgical procedure-related core treatment measure category are merged to obtain a database of real core treatment measure category.

[0084] Specifically, by merging the core treatment measures categories for medication and surgical procedures obtained from statistical analysis of medical record data, we can obtain a set of core treatment measures categories for each disease in the data, thus obtaining a database of true core treatment measures categories.

[0085] Furthermore, such as Figure 2 As shown, this figure is a flowchart of a method for combining disease phenotypes provided in an embodiment of this application. Based on medical record data and the core treatment item category database, diseases with the same core treatment item category under the same diagnosis are classified as the same disease phenotype, thus obtaining a disease phenotype classification set. This can be specifically implemented through steps S301 to S303.

[0086] S301: Based on medical record data and the core treatment measure category database, diseases with the same core treatment category under the same diagnosis are classified as the same disease phenotype, and the original disease phenotype classification set is obtained.

[0087] The original disease phenotype classification set may contain disease phenotypes with high similarity. In order to further improve the accuracy of disease phenotype classification, the original disease phenotype classification set can be optimized.

[0088] S302: Calculate the similarity between the various disease phenotypes in the original disease phenotype classification set.

[0089] S303: Merge the disease phenotypes with similarity higher than a preset value to obtain the disease phenotype classification set.

[0090] Specifically, since the original disease phenotype classification set may contain disease phenotypes with extremely high similarity, to further improve the accuracy of disease phenotype classification, after calculating the similarity between various disease phenotypes in the original disease phenotype classification set, disease phenotypes with similarity higher than a preset value can be merged to obtain a new disease phenotype classification set. The resulting disease phenotype classification set has stronger clinical guidance and can provide more accurate reference value for treatment plan formulation and prognosis.

[0091] Furthermore, the above-mentioned S302 can be implemented by S401 to S402.

[0092] S401: Generate a core treatment measure category vector for the disease phenotype based on the core treatment measure category corresponding to the disease phenotype in the original disease phenotype classification set.

[0093] For example, the core treatment item categories for a certain disease are (a, b, c, d, e), where a, b, c, d, and e are different core treatment item categories; α and β are two disease phenotypes in the disease phenotype classification set, where the core treatment item categories corresponding to disease phenotype α are a, b, c, and d, denoted as disease phenotype α(a, b, c, d); and the core treatment item categories corresponding to disease phenotype β are a, b, c, and e, denoted as disease phenotype β(a, b, c, e). A value containing the corresponding core treatment item category is denoted as 1, and a value not containing the corresponding core treatment item category is denoted as 0. This yields the core treatment item category vector for disease phenotype α as (1, 1, 1, 1, 0) and the core treatment item category vector for disease phenotype β as (1, 1, 1, 0, 1).

[0094] Based on this, weight values ​​can be assigned to the core treatment item categories in the core treatment item category vector of the disease phenotype, which can be specifically represented by equation (1):

[0095] IMPTNB = log(p(x) i =1|y j =1))-log(p(x i =1|y j =0)) (1)

[0096] Wherein, IMPTNB is the weight value, log(p(x) i =1|y j =1)) indicates the presence of disease phenotype y j Core treatment measures under the conditions, category x i The logarithmic probability value, log(p(x) i =1|y j =0)) indicates that the disease phenotype y does not appear. j Core treatment measures under the conditions, category x i The logarithmic probability value.

[0097] Specifically, the weight value can be understood as the probability of a certain core treatment category appearing under the condition of being diagnosed with a certain disease phenotype. For example, if 10 patients are diagnosed with disease phenotype y1, and 5 patients receive core treatment categories x1 and x2, while the other 5 patients only receive core treatment category x1, then the probability of x1 appearing in y1 is 100%, meaning the weight value of x1 for disease y1 is 1. Similarly, the weight value of x2 for disease phenotype y1 is 0.5. Each patient may correspond to multiple core treatment categories. The weight value of each core treatment category is related to the number of patients diagnosed with that disease phenotype and the number of patients among those diagnosed with that disease phenotype who received that core treatment category. In this application, the magnitude of the weight value can represent the importance of each core treatment category to a certain disease phenotype; the larger the weight value, the more important the corresponding core treatment category is to that disease phenotype.

[0098] For example, assuming that the weight values ​​of the core treatment item categories a, b, c, d, and e are all 0.5, then after assigning weight values, the core treatment item category vector for disease phenotype α is (1×0.5, 1×0.5, 1×0.5, 1×0.5, 0×0.5), and the core treatment item category vector for disease phenotype β is (1×0.5, 1×0.5, 1×0.5, 0×0.5, 1×0.5).

[0099] S402: Calculate the similarity between disease phenotypes based on the category vector of the core treatment measures corresponding to each disease phenotype.

[0100] Specifically, the similarity between disease phenotype α and disease phenotype β can be calculated using the cosine similarity formula, as shown in equation (2):

[0101]

[0102] Where A and B represent the two disease phenotype vectors for which similarity needs to be calculated. For example, A corresponds to disease phenotype α, and B corresponds to disease phenotype β. When the core treatment item category vector of disease phenotype α is (0.5, 0.5, 0.5, 0.5, 0) and the core treatment item category vector of disease phenotype β is (0.5, 0.5, 0.5, 0, 0.5), the similarity between disease phenotype α and disease phenotype β can be calculated as 0.75 by the above formula (2).

[0103] Furthermore, it also includes:

[0104] The names of the core treatment measures categories are standardized.

[0105] In summary, this application distinguishes disease phenotypes based on the categories of core treatment measures corresponding to core treatment measures, avoiding the influence of factors with little correlation to the disease during the classification process and improving the accuracy of disease phenotype classification. Furthermore, within the disease phenotype classification set, different disease phenotypes of the same disease can be distinguished. Since disease phenotypes usually have a more definite correlation with treatment decisions, the disease phenotype classification set obtained through the method provided in this application has stronger clinical guidance and can provide more accurate reference value for treatment plan formulation and prognosis.

[0106] This application also provides a device for disease phenotype classification, see [link to relevant documentation]. Figure 3 The figure is a schematic diagram of a device for classifying disease phenotypes according to an embodiment of this application. Its specific implementation and the technical effects achieved are consistent with those described in the embodiments of the above method, and some details will not be repeated here.

[0107] A device for classifying disease phenotypes, comprising:

[0108] The module consists of a generation module 1101, an establishment module 1102, and an acquisition module 1103;

[0109] The generation module 1101 is used to generate a database of core treatment measures for diseases based on the correspondence between each disease and each core treatment measure item in the knowledge source.

[0110] The establishment module 1102 is used to establish a core treatment measure item category database based on the core treatment measure item category corresponding to each core treatment measure item in the disease core treatment measure item database;

[0111] The obtaining module 1103 is used to obtain a disease phenotype classification set by classifying diseases with the same core treatment item category under the same diagnosis as the same disease phenotype based on medical record data and the core treatment item category database.

[0112] Further, the obtaining module 1103 includes:

[0113] Acquisition unit, calculation unit, and merging unit;

[0114] The obtaining unit is used to, based on medical record data and the core treatment measure category database, classify diseases with the same core treatment category under the same diagnosis as the same disease phenotype, and obtain the original disease phenotype classification set.

[0115] The computing unit is used to calculate the similarity between various disease phenotypes in the original disease phenotype classification set;

[0116] The merging unit is used to merge disease phenotypes with similarity higher than a preset value to obtain the disease phenotype classification set.

[0117] Furthermore, the computing unit includes:

[0118] Generation unit and computation subunit;

[0119] The generation unit is used to generate a core treatment measure category vector for the disease phenotype based on the core treatment measure category corresponding to the disease phenotype in the original disease phenotype classification set.

[0120] The computational subunit is used to calculate the similarity between disease phenotypes based on the category vector of the core treatment measures corresponding to each disease phenotype.

[0121] Further, the obtaining module 1103 includes:

[0122] Statistical unit and acquisition sub-unit;

[0123] The statistical unit is used to statistically analyze the core treatment measure categories corresponding to each disease in the medical record data to obtain a database of real core treatment measure category categories.

[0124] The obtaining subunit is used to classify diseases with the same core treatment item category under the same diagnosis as the same disease phenotype based on the real core treatment item category database and the core treatment item category database, thereby obtaining a disease phenotype classification set.

[0125] Furthermore, the statistical unit includes:

[0126] Statistical sub-units and merged sub-units;

[0127] The statistical subunit is used to statistically analyze the categories of core drug-related treatment measures and core surgical procedure-related treatment measures for each disease in the medical record data.

[0128] The merging subunit is used to merge the drug-related core treatment measure category with the surgical procedure-related core treatment measure category to obtain a database of real core treatment measure category.

[0129] Furthermore, it also includes:

[0130] The names of the core treatment measures categories are standardized.

[0131] In summary, the device provided in this application distinguishes disease phenotypes based on the categories of core treatment measures corresponding to those core treatment measures, avoiding the influence of factors with little correlation to the disease during the classification process and improving the accuracy of disease phenotype classification. Furthermore, within the disease phenotype classification set, different disease phenotypes of the same disease can be distinguished. Since disease phenotypes typically have a more definite correlation with treatment decisions, the disease phenotype classification set obtained through the method provided in this application has stronger clinical guidance value and can provide more accurate reference value for treatment plan formulation and prognosis.

[0132] This application also provides an electronic device, which includes a processor and a memory:

[0133] The memory is used to store program code and transmit the program code to the processor;

[0134] The processor is used to execute the steps of a method for classifying disease phenotypes as described above, according to the instructions in the program code.

[0135] like Figure 4 As shown, this figure is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. The electronic device 01 is represented in the form of a general electronic device. The components of the electronic device 01 may include, but are not limited to: one or more processors or processing units 03, system memory 08, and bus 04 connecting different system components (including system memory 08 and processing unit 03).

[0136] It is understood that the structure illustrated in this embodiment does not constitute a specific limitation on the electronic device. In other embodiments, the electronic device may include more or fewer components than illustrated, or combine some components, or split some components, or have different component arrangements. The illustrated components may be implemented in hardware, software, or a combination of software and hardware.

[0137] Bus 04 represents one or more of several bus architectures, including memory buses or memory controllers, peripheral buses, graphics acceleration ports, processors, or local buses using any of the various bus architectures. Examples of these architectures include, but are not limited to, the Industry Standard Architecture (ISA) bus, the Micro Channel Architecture (MAC) bus, the Enhanced ISA bus, the Video Electronics Standards Association (VESA) local bus, and the Peripheral Component Interconnect (PCI) bus.

[0138] System memory 08 may include computer-readable storage media in the form of volatile memory, such as random access memory (RAM) 09 and / or cache memory 10. Electronic device 01 may further include other removable / non-removable, volatile / non-volatile computer system storage media. By way of example only, storage system 11 may be used to read and write non-removable, non-volatile magnetic media (…). Figure 4 Not shown; usually referred to as a "hard drive"). Although Figure 4 Not shown, a disk drive for reading and writing to a removable non-volatile disk (e.g., a "floppy disk") and an optical disk drive for reading and writing to a removable non-volatile optical disk (e.g., a CD-ROM, DVD-ROM, or other optical media) may be provided. In these cases, each drive may be connected to bus 04 via one or more data media interfaces. Memory 08 may include at least one program product having a set (e.g., at least one) of program modules configured to perform the functions of the embodiments of the present invention.

[0139] A program / utility 12 having a set (at least one) of program modules 13 may be stored in, for example, memory 08. Such program modules 13 include, but are not limited to, an operating system, one or more application programs, other program modules, and program data. Each or some combination of these examples may include an implementation of a network environment. Program modules 13 typically perform the functions and / or methods described in the embodiments of the present invention.

[0140] Electronic device 01 can also communicate with one or more external devices 02 (e.g., keyboard, pointing device, display 07, etc.), and with one or more devices that enable a user to interact with the electronic device 01, and / or with any device that enables the electronic device 01 to communicate with one or more other computing devices (e.g., network card, modem, etc.). This communication can be performed through input / output (I / O) interface 06. Furthermore, electronic device 01 can also communicate with one or more networks (e.g., local area network (LAN), wide area network (WAN), and / or public networks, such as the Internet) through network adapter 05. Figure 4As shown, network adapter 05 communicates with other modules of electronic device 01 via bus 04. It should be understood that, although... Figure 4 As not shown, other hardware and / or software modules may be used in conjunction with electronic device 01, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage systems.

[0141] The processor unit 03 executes various functional applications and data processing by running programs stored in the system memory 08, such as implementing the disease phenotype classification method provided in the embodiments of this application.

[0142] This application also provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the steps of a method for classifying disease phenotypes as described above.

[0143] It should be noted that the various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referred to mutually. Each embodiment focuses on describing the differences from other embodiments. In particular, for the embodiments of apparatus, devices, and storage media, since they are basically similar to the method embodiments, the description is relatively simple, and the relevant parts can be referred to the description of the method embodiments. The apparatus, devices, and storage media embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate, and the components indicated as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of the solution in this embodiment according to actual needs. Those skilled in the art can understand and implement this without creative effort.

[0144] The above description is merely one 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 technical scope 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 classifying disease phenotypes, characterized in that, include: Based on the correspondence between each disease and each core treatment measure item in the knowledge source, a database of core treatment measures items for diseases is generated; Based on the core treatment measure item database, a core treatment measure item category database is established for each core treatment measure item corresponding to the core treatment measure item category. Based on medical record data and the database of core treatment measures categories, diseases with the same core treatment measures category under the same diagnosis are classified as the same disease phenotype, thus obtaining a disease phenotype classification set. Based on medical record data and the core treatment item category database, diseases with the same core treatment item category under the same diagnosis are classified as the same disease phenotype, thus obtaining a disease phenotype classification set, including: Based on medical record data and the database of core treatment measures, diseases with the same core treatment measures category under the same diagnosis are classified as the same disease phenotype, thus obtaining the original disease phenotype classification set. Calculate the similarity between each disease phenotype in the original disease phenotype classification set; The disease phenotypes with similarity higher than a preset value are merged to obtain the disease phenotype classification set; or, By statistically analyzing the core treatment measures categories corresponding to each disease in the medical record data, a database of real core treatment measures categories is obtained. Based on the database of actual core treatment measures and the database of core treatment measures, diseases with the same core treatment measures category under the same diagnosis are classified as the same disease phenotype, thus obtaining a disease phenotype classification set.

2. The method according to claim 1, characterized in that, The calculation of the similarity between various disease phenotypes in the original disease phenotype classification set includes: Based on the core treatment item category corresponding to the disease phenotype in the original disease phenotype classification set, generate the core treatment item category vector of the disease phenotype. The similarity between disease phenotypes is calculated based on the category vector of the core treatment measures corresponding to each disease phenotype.

3. The method according to claim 1, characterized in that, The core treatment measure categories corresponding to each disease in the statistical medical record data are used to obtain a database of real core treatment measure category categories, including: The categories of core drug-related treatment measures and core surgical procedure-related treatment measures for each disease in the statistical medical record data are analyzed. The drug-related core treatment measure category and the surgical procedure-related core treatment measure category are merged to obtain a database of real core treatment measure category.

4. The method according to any one of claims 1-3, characterized in that, Also includes: The names of the core treatment measures categories are standardized.

5. A device for classifying disease phenotypes, characterized in that, include: Module generation, module creation, and module acquisition; The generation module is used to generate a database of core treatment measures for diseases based on the correspondence between each disease and each core treatment measure item in the knowledge source. The establishment module is used to establish a core treatment measure item category database based on the core treatment measure item category corresponding to each core treatment measure item in the disease core treatment measure item database; The obtaining module is used to, based on medical record data and the core treatment measure category database, classify diseases with the same core treatment measure category under the same diagnosis as the same disease phenotype, and obtain a disease phenotype classification set. The obtaining module includes: Acquisition unit, calculation unit, and merging unit; The obtaining unit is used to, based on medical record data and the core treatment measure category database, classify diseases with the same core treatment measure category under the same diagnosis as the same disease phenotype, and obtain the original disease phenotype classification set. The computing unit is used to calculate the similarity between various disease phenotypes in the original disease phenotype classification set; The merging unit is used to merge disease phenotypes with similarity higher than a preset value to obtain the disease phenotype classification set; or, The obtaining module includes: a statistics unit and an obtaining subunit; The statistical unit is used to statistically analyze the core treatment measure categories corresponding to each disease in the medical record data to obtain a database of real core treatment measure category categories. The obtaining subunit is used to classify diseases with the same core treatment item category under the same diagnosis as the same disease phenotype based on the real core treatment item category database and the core treatment item category database, thereby obtaining a disease phenotype classification set.

6. An electronic device, characterized in that, The electronic device includes a processor and a memory: The memory is used to store program code and transmit the program code to the processor; The processor is configured to execute the steps of a method for classifying disease phenotypes as described in any one of claims 1-4, according to instructions in the program code.

7. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, implements the steps of a method for classifying disease phenotypes as described in any one of claims 1-4.

Images