Disease differential diagnosis method and related device

Abstract

The invention discloses a disease differential diagnosis method and a related device, and the method comprises the steps: building a core identification database through an obtained disease differential diagnosis database and a disease phenotype classification database, and carrying out the matching of a target disease phenotype corresponding to a target disease of an inquiry user with the core identification database, obtaining a target core identification item corresponding to the target disease phenotype, and then determining an identification diagnosis result of the interrogation user according to the target core identification item. Wherein a core identification database between the disease phenotype and the core identification item is established according to the corresponding relation between the disease of the disease identification database and the core identification item and the corresponding relation between the disease phenotype and the disease in the disease phenotype classification database, and the core identification item is determined according to the disease phenotype of the patient and the core identification database. The core identification item comprises at least one to-be-diagnosed disease, so that the identification and diagnosis result is precisely subdivided into the to-be-identified disease associated with the target disease, and the precision of the identification and diagnosis result is improved.

Description

Technical Field

[0001] The present application relates to the field of computer technology, and in particular to a disease differential diagnosis method and related devices. Background Art

[0002] At present, when patients visit doctors, doctors generally make differential diagnosis of diseases based on the patient's self-report and the patient's medical examination results. Differential diagnosis refers to the process of excluding other possible diseases or conditions through analysis and comparison based on a series of symptoms, signs or examination results presented by the patient, and finally determining the diagnosis of the most likely disease of the patient.

[0003] However, this method is limited by the doctor's professional knowledge, experience level and the accuracy of medical examination results, which may lead to misdiagnosis and missed diagnosis, and further delay the timeliness of patient treatment. In order to solve this technical problem, a disease diagnosis model based on machine learning is provided in the related technology. The disease diagnosis model is obtained by machine learning training based on medical big data, which can assist doctors in differential diagnosis of patients' diseases.

[0004] However, the current disease diagnosis model only supports a certain type of disease as a differential diagnosis result, which may be the same as the doctor's initial differential diagnosis result, and cannot provide substantive advice to doctors. The accuracy of the disease differential diagnosis results is low. Summary of the invention

[0005] Based on the above problems, the present application provides a disease differential diagnosis method and related devices, aiming to solve the problem of low accuracy of differential diagnosis results.

[0006] The embodiments of the present application disclose the following technical solutions:

[0007] In a first aspect, an embodiment of the present application provides a method for differential diagnosis of a disease, the method comprising:

[0008] Acquire a disease differential diagnosis database and a disease phenotype classification database; the disease differential diagnosis database includes the correspondence between each disease and each core identification item, the core identification item includes at least one disease to be identified associated with the corresponding disease; the disease phenotype classification database includes the correspondence between each disease phenotype and each disease;

[0009] A core identification database is established based on the disease differential diagnosis database and the disease phenotype classification database; the core identification database includes the corresponding relationship between each disease phenotype and each core identification item; the corresponding relationship between each disease phenotype and each core identification item is determined by each disease;

[0010] Matching the target disease phenotype corresponding to the target disease of the inquiring user with the core identification database to obtain the target core identification item corresponding to the target disease phenotype;

[0011] Determine the differential diagnosis result of the consulting user based on the target core identification items.

[0012] Optionally, determining the differential diagnosis result of the user inquiring according to the target core identification item includes:

[0013] Determining an actual core identification item corresponding to the target disease; the actual core identification item includes at least one actual disease to be identified associated with the target disease;

[0014] The actual to-be-identified disease in the actual core identification item that is identical to at least one to-be-identified disease in the target core identification item is used as the differential diagnosis result of the inquiring user.

[0015] Optionally, when the target disease has multiple corresponding target disease phenotypes, the method further comprises:

[0016] Obtain target core identification items corresponding to each target disease phenotype;

[0017] Calculating the similarity between every two target core identification items in the plurality of target core identification items;

[0018] Merging two target core identification items corresponding to the target similarities among the multiple similarities to obtain a merged core identification item; the target similarity is a similarity among the multiple similarities that is greater than a preset similarity;

[0019] The unmerged target core identification items and the merged core identification items among the multiple target core identification items are used as the differential diagnosis results of the inquiring user.

[0020] Optionally, calculating the similarity between every two target core identification items in the plurality of target core identification items comprises:

[0021] Acquire an actual core identification item corresponding to the target disease; the actual core identification item includes at least one actual disease to be identified associated with the target disease;

[0022] Determine a target vector for each target core identification item according to the actual core identification item; the target vector includes a first element and a second element, the first element includes the disease to be identified in the target core identification item that is the same as the actual disease to be identified in the actual core identification item, and the second element includes the disease to be identified in the target core identification item that is different from the actual disease to be identified in the actual core identification item;

[0023] The similarity between each two target core discriminants is calculated based on each target vector.

[0024] Optionally, before determining the target vector of each target core identification item according to the actual core identification item, the method further comprises:

[0025] Standardizing the actual diseases to be identified in the actual core identification items to obtain standard actual core identification items; and standardizing each target core identification item to obtain multiple standard target core identification items;

[0026] Determining a target vector for each target core identification item based on the actual core identification item includes: determining a target vector for each standard target core identification item based on the standard actual core identification item.

[0027] Optionally, the disease differential diagnosis database is obtained in the following manner: generating the disease differential diagnosis database according to the correspondence between each disease in the knowledge source and each differential diagnosis; the differential diagnosis includes the diseases to be identified in the diagnosis process of the corresponding disease.

[0028] Optionally, the disease phenotype classification database is obtained by:

[0029] Acquire multiple medical record data; the medical record data includes disease information of corresponding historical consultation users;

[0030] Processing multiple medical record data to obtain multiple standard medical record data;

[0031] Classify multiple standard medical record data based on the pre-trained disease phenotype classification model;

[0032] A disease phenotype classification database is established based on classified multiple standard medical record data.

[0033] In a second aspect, an embodiment of the present application provides a disease differential diagnosis device, the device comprising:

[0034] An acquisition module is used to acquire a disease differential diagnosis database and a disease phenotype classification database; the disease differential diagnosis database includes a correspondence between each disease and each core identification item, and the core identification item includes at least one disease to be identified associated with the corresponding disease; the disease phenotype classification database includes a correspondence between each disease phenotype and each disease;

[0035] The core identification database establishment module is used to establish a core identification database based on the disease differential diagnosis database and the disease phenotype classification database; the core identification database includes the corresponding relationship between each disease phenotype and each core identification item; the corresponding relationship between each disease phenotype and each core identification item is determined by each disease;

[0036] A target core identification item determination module is used to match the target disease phenotype corresponding to the target disease of the consulting user with the core identification database to obtain the target core identification item corresponding to the target disease phenotype;

[0037] The differential diagnosis module is used to determine the differential diagnosis results of the consulting user based on the target core identification items.

[0038] In a third aspect, an embodiment of the present application provides a computer device, comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, a disease differential diagnosis method as described in any one of the first aspects is implemented.

[0039] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, wherein instructions are stored in the computer-readable storage medium. When the instructions are executed on a terminal device, the terminal device executes a disease differential diagnosis method as described in any one of the first aspects.

[0040] Compared with the prior art, this application has the following beneficial effects:

[0041] The disease differential diagnosis method provided in the present application establishes a core identification database based on the acquired disease differential diagnosis database and disease phenotype classification database; wherein the core identification database includes the correspondence between each disease phenotype and each core identification item, the disease differential diagnosis database includes the correspondence between each disease and each core identification item, the core identification item includes at least one disease to be identified associated with the corresponding disease, the disease phenotype classification database includes the correspondence between each disease phenotype and each disease, and the correspondence between each disease phenotype and each core identification item is determined by each disease; further, the target disease phenotype corresponding to the target disease of the consulting user is matched with the core identification database to obtain the target core identification item corresponding to the target disease phenotype, and then the differential diagnosis result of the consulting user is determined according to the target core identification item.

[0042] Among them, the correspondence between disease phenotypes and core identification items is established through the correspondence between diseases and core identification items in the disease identification database, and the correspondence between disease phenotypes and diseases in the disease phenotype classification database (core identification database), and then the core identification items are determined through the patient's disease phenotype and the core identification database. The core identification items include at least one disease to be diagnosed, so that the differential diagnosis results are accurately subdivided into diseases to be identified that are associated with the target disease, thereby improving the accuracy of the differential diagnosis results. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0044] Figure 1 A flowchart of a disease differential diagnosis method provided in an embodiment of the present application;

[0045] Figure 2 A schematic diagram of another method for differential diagnosis of diseases provided in an embodiment of the present application;

[0046] Figure 3 A schematic diagram of a process for determining a disease phenotype classification set provided in an embodiment of the present application;

[0047] Figure 4 A schematic diagram of a process flow of a disease differential diagnosis device provided in an embodiment of the present application;

[0048] Figure 5 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0049] At present, when patients visit the doctor, doctors generally make differential diagnosis of diseases based on the patient's self-report and the patient's medical examination results. Differential diagnosis refers to the process of excluding other possible diseases or conditions through analysis and comparison based on a series of symptoms, signs or examination results presented by the patient, and finally determining the diagnosis of the patient's most likely disease. As an example, assuming that the doctor initially determines that the patient's disease is anemia based on the patient's condition, the doctor also needs to distinguish whether it is aplastic anemia or nutritional anemia, and exclude the possibility that the anemia is hemolytic anemia. This process is called differential diagnosis, and the identification of possible aplastic anemia and nutritional anemia is the result of differential diagnosis.

[0050] However, this method is limited by the doctor's professional knowledge, experience level and the accuracy of medical examination results, which may lead to misdiagnosis and missed diagnosis, and further delay the timeliness of patient treatment. In order to solve this technical problem, a disease diagnosis model based on machine learning is provided in the related technology. The disease diagnosis model is obtained by machine learning training based on medical big data, which can assist doctors in differential diagnosis of patients' diseases.

[0051] However, the current disease diagnosis model only supports a certain type of disease as a differential diagnosis result, which may be the same as the doctor's preliminary differential diagnosis result, and cannot provide substantive advice to doctors. The accuracy of the disease differential diagnosis result is low. As an example, the result that the disease diagnosis model can support is anemia as a differential diagnosis result, but it cannot further subdivide whether it is nutritional anemia or hemolytic anemia as a differential diagnosis result. However, the doctor can also preliminarily diagnose that the patient's disease is mainly anemia.

[0052] In order to solve the above problems, the embodiment of the present application provides a disease differential diagnosis method and related devices. The method includes: establishing a core identification database based on the acquired disease differential diagnosis database and disease phenotype classification database; wherein the core identification database includes the corresponding relationship between each disease phenotype and each core identification item, the disease differential diagnosis database includes the corresponding relationship between each disease and each core identification item, the core identification item includes at least one disease to be identified associated with the corresponding disease, the disease phenotype classification database includes the corresponding relationship between each disease phenotype and each disease, and the corresponding relationship between each disease phenotype and each core identification item is determined by each disease; further matching the target disease phenotype corresponding to the target disease of the consulting user with the core identification database, obtaining the target core identification item corresponding to the target disease phenotype, and then determining the differential diagnosis result of the consulting user according to the target core identification item.

[0053] In this way, through the correspondence between diseases and core identification items in the disease identification database, and the correspondence between disease phenotypes and diseases in the disease phenotype classification database, a correspondence between disease phenotypes and core identification items (core identification database) is established, and then the core identification items are determined through the patient's disease phenotype and the core identification database. The core identification items include at least one disease to be diagnosed. In this way, the differential diagnosis results are accurately subdivided into diseases to be identified that are associated with the target disease, thereby improving the accuracy of the differential diagnosis results.

[0054] In order to enable those skilled in the art to better understand the solution of the present application, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0055] See also Figure 1 , which is a flow chart of a disease differential diagnosis method provided in an embodiment of the present application, combined with Figure 1 As shown, the disease differential diagnosis method provided in the embodiment of the present application may include:

[0056] S101: Acquire a disease differential diagnosis database and a disease phenotype classification database.

[0057] The disease differential diagnosis database includes the correspondence between each disease and each core identification item, and the core identification item includes at least one disease to be identified associated with the corresponding disease. The core identification item means the differential diagnosis category of the corresponding disease, and the differential diagnosis category includes at least one disease to be identified. It should be understood that generally each disease will have a disease category or disease with similar symptoms, so the disease category or disease to be identified is the differential diagnosis category (or core identification item).

[0058] As an example, if the disease is anemia, the core identification items corresponding to anemia may be iron deficiency anemia, vitamin B12 deficiency anemia, and folate deficiency anemia; if the disease is migraine, the core identification items may be tension headache, cluster headache, intracranial space-occupying lesions (such as tumors), intracranial infections (such as meningitis), and cerebrovascular diseases (such as cerebral vasospasm). Among them, intracranial positional lesions, intracranial infections, and cerebrovascular diseases belong to the general category of diseases.

[0059] In a possible implementation, a disease differential diagnosis database is generated based on the correspondence between each disease and each differential diagnosis in the knowledge source. Among them, the differential diagnosis includes the diseases to be identified in the diagnosis process of the corresponding disease. Specifically, the knowledge source refers to materials such as medical textbooks, literature, clinical guidelines and national relevant documents. The knowledge source is used as a reliable basis for determining the correspondence between each disease and each differential diagnosis. In addition, standards such as the International Classification of Diseases (ICD-10) and / or the Unified Medical Language System (UMLS) can also be used to standardize diseases and differential diagnoses to avoid the problem of non-professional terminology or inconsistent professional terminology.

[0060] The disease phenotype classification database includes the correspondence between each disease phenotype and each disease. Disease phenotype means the manifestation or characteristics of the disease, such as the disease phenotype of anemia is pale complexion, and the disease phenotype of migraine is frequent unilateral headache and vomiting. It should be understood that the disease phenotype classification database records the phenotypic classification of various diseases, that is, the external manifestations or characteristics of the disease. These phenotypic classifications may be associated with specific diseases or disease groups.

[0061] As a possible implementation method, you can first obtain multiple medical record data including the disease information of the corresponding historical consultation users, then process the multiple medical record data to obtain multiple standard medical record data, and then classify the multiple standard medical record data according to the pre-trained disease phenotype classification model, and establish a disease phenotype classification database based on the classified multiple standard medical record data.

[0062] Medical record data refers to the user's current and previous medical information data, which may specifically include disease information. Disease information refers to information related to the disease suffered by the user, such as test results, self-described condition, doctor's inquiry results, doctor's preliminary diagnosis results, etc. Standard medical record data refers to processed medical record data. It should be noted that due to the large amount of medical record data and inconsistent terminology, it may lead to incorrect classification of disease phenotypes. Therefore, the medical record data needs to be processed to obtain standard case data. The pre-trained disease phenotype classification model refers to a model used to classify various diseases. The pre-trained disease phenotype classification model can also be integrated into a disease phenotype classification system, which is not specifically limited here.

[0063] Since the medical record data reflects the performance or representation of the consulting user, on the basis of determining the disease corresponding to the medical record data, the disease phenotype corresponding to the disease can be determined through a pre-trained disease phenotype classification model.

[0064] It should be noted that the relationship between diseases and disease phenotypes in the disease phenotype classification library can be one-to-one, one-to-many, or many-to-one. For example, the disease phenotype of anemia is pale complexion, and the disease phenotype of migraine is frequent unilateral headache and vomiting. Specifically, one disease corresponds to one disease phenotype, and one disease phenotype may have multiple manifestations or characteristics.

[0065] S102: Establish a core identification database based on the disease differential diagnosis database and the disease phenotype classification database.

[0066] The core identification database includes the correspondence between each disease phenotype and each core identification item. The correspondence between each disease phenotype and each core identification item is determined by each disease. For example, assuming that the disease is anemia, the disease phenotype of anemia is pale complexion and weakness in the limbs, and the core identification items of anemia are nutritional anemia, aplastic anemia, and iron deficiency anemia, then in the established core identification database, the core identification items corresponding to pale complexion and weakness in the limbs are nutritional anemia, aplastic anemia, and iron deficiency anemia.

[0067] It should be noted that since the doctor's preliminary diagnosis may have certain deviations, in order to avoid errors in the differential diagnosis results caused by only correlating the core identification items corresponding to the disease association, in the embodiment of the present application, by associating the disease phenotype with the core identification items, the core identification items can be determined by the patient's disease phenotype, eliminating the influence of subjective factors, and making the differential diagnosis results more accurate. As an example, assuming that the doctor's preliminary diagnosis result is anemia, and the disease phenotype is pale complexion and toothache, then the actual disease may be neuralgic toothache, not anemia.

[0068] S103: Match the target disease phenotype corresponding to the target disease of the consulting user with the core identification database to obtain the target core identification item corresponding to the target disease phenotype.

[0069] The target disease refers to the disease suffered by the consulting user. The target disease refers to a hypothetical disease or a disease that is initially diagnosed by a doctor. A hypothetical disease refers to a disease that is difficult for a doctor to determine based on the disease phenotype. In this case, the target disease can be a disease assumed by the doctor based on experience, or the disease is unclear and only the disease phenotype is given. The target disease phenotype is the disease phenotype corresponding to the target disease. The target disease phenotype can be obtained based on the consulting user's medical record data, self-reported condition, doctor's observation, and doctor's inquiry. The method of obtaining the target disease phenotype is not limited here.

[0070] The target core identification item refers to the core identification item in the core identification database that matches the target disease phenotype. For example, the target disease phenotype is pale complexion and weakness in the limbs. The core identification database includes disease phenotype A as pale complexion and weakness in the limbs. The core identification items corresponding to disease phenotype A are nutritional anemia, aplastic anemia and iron deficiency anemia. Then the target core identification items are nutritional anemia, aplastic anemia and iron deficiency anemia.

[0071] It should be understood that in the embodiment of the present application, by matching the target disease phenotype of the user in question with the disease phenotype in the core identification database, the target core identification item corresponding to the target disease phenotype is determined, and the differential diagnosis of different disease phenotypes of the same disease can be distinguished by disease phenotype, avoiding the influence of some differential diagnoses that are not closely related to the disease in the diagnosis process, and improving the accuracy of the differential diagnosis results. At the same time, it can avoid the problem of deviation in the differential diagnosis results caused by errors in the initial diagnosis of the doctor's disease, and improve the accuracy of the differential diagnosis results.

[0072] S104: Determine the differential diagnosis result of the consulting user according to the target core identification item.

[0073] It should be understood that after determining the target core identification item, the differential diagnosis result corresponding to the target disease of the target consulting user can be determined based on the target core identification item. In this way, the differential diagnosis result can be accurately subdivided into the diseases to be identified associated with the target disease, thereby improving the accuracy of the differential diagnosis result.

[0074] In addition, it can also avoid the deviation of differential diagnosis results caused by errors in the disease diagnosed in the preliminary diagnosis results, thereby improving the accuracy of differential diagnosis results.

[0075] Based on the disease differential diagnosis method provided in the above embodiment, in some possible implementations, step S104 may specifically include:

[0076] A1: Determine the actual core discriminant items corresponding to the target disease.

[0077] The actual core identification item includes at least one actual disease to be identified that is associated with the target disease.

[0078] It should be understood that the medical record data of the consulting user may contain a large amount of diagnosis-related information, such as doctor's diagnosis, symptoms, examinations and tests, previous medical history, etc. The actual core identification items can be determined based on the case data of the consulting user, that is, the core identification items corresponding to the target disease in the actual consultation environment.

[0079] A2: The actual diseases to be identified in the actual core identification items that are the same as at least one disease to be identified in the target core identification items are used as the differential diagnosis results of the inquiring user.

[0080] After determining the actual core identification item of the target disease, the actual core identification item is matched with the target core identification item to use the same diseases to be identified between the target core identification item and the actual core identification item as the differential diagnosis results. As an example, assuming that the target core identification items are iron deficiency anemia and vitamin B12 deficiency anemia, and the actual core identification items are iron deficiency anemia, vitamin B12 deficiency anemia, and aplastic anemia, by matching the target core identification item with the actual core identification item, the differential diagnosis results can be determined to be iron deficiency anemia and vitamin B12 deficiency anemia, eliminating the interference of aplastic anemia and improving the accuracy of the differential diagnosis results.

[0081] It should be understood that this embodiment can further eliminate the interference of non-critical diseases to be identified, thereby ensuring the accuracy and reliability of the differential diagnosis results.

[0082] Based on the disease differential diagnosis method provided in the above embodiment, when the target disease has multiple corresponding target disease phenotypes, the method may further include:

[0083] B1: Obtain the target core identification items corresponding to each target disease phenotype.

[0084] It should be noted that the target disease phenotype can be at least one manifestation or feature. Therefore, in some possible implementations, each manifestation or feature can be used as a target disease phenotype separately. For example, if the target disease is anemia, the manifestation or feature corresponding to anemia is dizziness, memory loss, sleep disorder, lethargy, and coma, then each manifestation can be used as a target disease phenotype. It is also possible to group manifestations or features according to different categories, and use multiple manifestations or features in the same group as a target disease phenotype. For example, if the target disease is anemia, the manifestation or feature corresponding to anemia is dizziness, memory loss, sleep disorder, lethargy, and coma, then the aforementioned manifestations or features can be divided into dizziness and memory loss, sleep disorder, lethargy, and coma, and the three groups of dizziness and memory loss, sleep disorder, lethargy, and coma after grouping are used as three target disease phenotypes. It is also possible to obtain multiple target disease phenotypes under different visiting doctors or visiting times based on medical record data, and each target disease phenotype can include multiple manifestations or features, such as the target disease phenotypes of the first visit are a and b, and the target disease phenotypes of the second visit are a, b, and c.

[0085] B2: Calculate the similarity between every two target core identification items in the multiple target core identification items.

[0086] Similarity refers to the degree of similarity between two target core discriminants.

[0087] It should be understood that when there are multiple corresponding target disease phenotypes for a target disease, since the target core identification items corresponding to the multiple target disease phenotypes are similar or identical, in order to improve the accuracy of the differential diagnosis result, it is necessary to merge the target core identification items with higher similarity or the sameness among the multiple target core identification items. Specifically, the similarity between each two target core identification items can be calculated, so that the two target core identification items with higher similarity can be merged.

[0088] As a possible implementation manner, step B2 may specifically include:

[0089] C1: Get the actual core discriminant items corresponding to the target disease.

[0090] The actual core identification item includes at least one actual disease to be identified associated with the target disease. It should be understood that in order to ensure that the core identification item is more appropriate to the actual core identification item in the actual medical consultation scenario, the actual core identification item can be used as a standard to vectorize the target core identification item.

[0091] As a possible implementation method, the actual diseases to be identified in the actual core identification items can also be standardized to obtain standard actual core identification items, and each target core identification item can be standardized to obtain multiple standard target core identification items. It should be understood that since the names of the diseases to be identified in the actual core identification items and the target core identification items may be inconsistent, in order to avoid inaccurate subsequent vectorization results, the diseases to be identified in the actual core identification items and the target core identification items can be standardized first to facilitate vectorization. Correspondingly, the following step C2 can be: determine the target vector of each standard target core identification item based on the standard actual core identification item.

[0092] C2: Determine the target vector of each target core discriminant based on the actual core discriminant.

[0093] Among them, the target vector includes a first element and a second element, the first element includes the diseases to be identified in the target core identification item that are the same as the actual diseases to be identified in the actual core identification item, and the second element includes the diseases to be identified in the target core identification item that are different from the actual diseases to be identified in the actual core identification item. It should be understood that the diseases to be identified in the target core identification item that are the same as the actual core identification item are recorded as the first element, such as 1, a, etc., and the diseases to be identified in the target core identification item that are different from the actual core identification item are recorded as the second element, such as 0, b, etc., to determine whether each target core identification item contains the diseases to be identified contained in the actual core identification item.

[0094] C3: Calculate the similarity between each two target core discriminants based on each target vector.

[0095] After determining the target vector corresponding to each target core identification item, the similarity between the two target core identification items can be calculated based on the target vector. Vectorization processing can facilitate calculation and reduce calculation complexity.

[0096] B3: Merge two target core identification items corresponding to the target similarities in the multiple similarities to obtain a merged core identification item.

[0097] The target similarity is a similarity greater than a preset similarity among multiple similarities. Merging core identification items means merging the results of the diseases to be identified in the two target core identification items corresponding to the target similarity. As an example, assuming that the target core identification item A includes a, b, c, d, and the target core identification item B includes a, b, c, e, then the merged core identification item may include a, b, c, d, e.

[0098] B4: Use the unmerged target core identification items and the merged core identification items among the multiple target core identification items as the differential diagnosis results of the inquiring user.

[0099] It should be understood that, since there may be a situation where two target core identification items are not similar, when determining the differential diagnosis result of the consulting user, it is necessary to use the unmerged target core identification item and the merged core identification item together as the differential diagnosis result.

[0100] Based on the disease differential diagnosis method provided in the above embodiment, see Figure 2 , which is a flow chart of another disease differential diagnosis method provided in the embodiment of the present application. Figure 2 As shown, the disease differential diagnosis method provided in the embodiment of the present application may include:

[0101] S201: Generate a disease differential diagnosis database based on the correspondence between each disease and differential diagnosis in the knowledge source.

[0102] It should be understood that before the disease suffered by the consulting user is diagnosed, it needs to be differentiated from other diseases before the final diagnosis can be made. During this period, the disease that needs to be differentiated is the differential diagnosis result. For example, if the preliminary diagnosis result is anemia, it may be necessary to distinguish it from nutritional anemia, aplastic anemia and hemolytic anemia. Nutritional anemia, aplastic anemia and hemolytic anemia are the differential diagnosis results.

[0103] In actual medical consultation scenarios, there is a definite correlation between differential diagnosis and diagnosis of diseases. Diagnosis is the doctor's judgment or conclusion on the disease based on the patient's medical history, symptoms, signs, and necessary laboratory and imaging examinations. Differential diagnosis is the process of further confirming or revising the diagnosis by excluding other possible diseases or conditions on the basis of the preliminary diagnosis. Diagnosis is the premise and basis of differential diagnosis. Without a preliminary diagnosis, targeted differential diagnosis cannot be carried out. Differential diagnosis is a deepening and refinement of diagnosis, which helps doctors to more accurately determine the nature and extent of the disease.

[0104] In actual scenarios, different disease phenotypes usually represent different characteristics or changes of the disease in the patient's body, which may affect the diagnosis, treatment and management of the disease. Therefore, when the disease phenotype of the patient's disease is different, the disease to be identified may also be different.

[0105] The differential diagnosis result can be a disease or a class of diseases, such as the differential diagnosis result as a class of diseases, such as infectious diseases, inflammatory diseases, immunodeficiency diseases, etc., or the differential diagnosis result as a specific disease, such as iron deficiency anemia.

[0106] It should be noted that, in the embodiment of the present application, the method for determining the disease differential diagnosis database can refer to the relevant description in step S101 in the above embodiment, and will not be repeated here.

[0107] S202: Determine core identification items corresponding to each disease phenotype based on the disease differential diagnosis database and the disease phenotype classification database, and establish a core identification database.

[0108] Among them, each core identification item can be standardized according to the relevant standard data set. For example: International Classification of Diseases Surgery and Procedure Codes (ICD-9-CM3). For example, the secondary classification of anemia is nutritional anemia, hemolytic anemia, aplastic and other anemias. The tertiary classification of nutritional anemia is iron deficiency anemia, vitamin B12 deficiency anemia, and folate deficiency anemia. In order to ensure the consistency of the name, the diseases to be identified in the core identification items can be standardized according to the third-level classification in ICD. That is, the third-level name in ICD is used as the unified name of the disease to be identified in the core identification items.

[0109] It should be noted that the method for determining the disease phenotype classification database is the same as the method for obtaining the disease phenotype root database in step S101 in the above embodiment, which will not be repeated here. At the same time, the method for obtaining the core identification database in step S202 in the embodiment of the present application is the same as the method for determining the core identification database in step S102 in the above embodiment. For relevant explanations, please refer to the explanation of step S102 in the above embodiment, which will not be repeated here.

[0110] S203: Determine a target core identification item set corresponding to a target disease phenotype corresponding to the target disease of the consulting user based on the medical record data and the core identification database.

[0111] Medical record data can come from records of actual diagnosis and treatment processes, such as the sum of text, charts, images and other materials generated by medical staff during medical activities.

[0112] As an example, taking fever as an example, it is assumed that the phenotypes corresponding to the disease include fever phenotype A and fever phenotype B. Fever phenotype A: persistent high fever, accompanied by chills and headache, and the core identification items corresponding to fever phenotype A are common cold, influenza, bacterial pneumonia, and meningitis; Fever phenotype B: low fever, accompanied by cough and fatigue, and the core identification items corresponding to fever phenotype B are influenza and common cold; Fever phenotype C: low fever in the afternoon, accompanied by night sweats and weight loss, and the core identification items of fever phenotype C are active tuberculosis and malignant tumors.

[0113] Based on the above example, assume that the core identification items corresponding to fever phenotypes A and B may overlap in some cases (such as immunocompromised patients), for example, both include "common cold, influenza". In this case, the core identification items corresponding to these two fever phenotypes can be classified into the same target core identification item set, that is, the phenotypes corresponding to the disease include fever phenotypes A and B, and their core identification items include common cold, influenza, bacterial pneumonia, and meningitis.

[0114] It should be understood that, based on the medical record data and the core identification database, the disease phenotypes corresponding to different core identification items can be distinguished; when the core identification items corresponding to the disease phenotypes are the same or similar, the core identification items corresponding to the various disease phenotypes can be merged to obtain a target core identification item set. This implementation can provide doctors with more diagnostic references.

[0115] As an example, assume that the core identification items corresponding to disease phenotype A are a, b, and c, and the core identification items corresponding to disease phenotype B are b, c, and d. Disease phenotype A and disease phenotype B have the same core identification items, but there are differences between the disease phenotype corresponding to core identification item A and the disease phenotype corresponding to core identification item B. Therefore, when the similarity between disease phenotype A and disease phenotype B is greater than a preset value, the core identification items corresponding to the two disease phenotypes are merged to obtain the target core identification item set, which is a, b, c, and d.

[0116] As a possible implementation, step S203 may be implemented by the following steps:

[0117] Step 1: Obtain the real core identification database based on the core identification items corresponding to the disease phenotypes in the medical record data.

[0118] The real core identification database reflects the correspondence between the disease phenotype and the core identification items in the real medical scene. It should be understood that in the embodiment of the present application, considering that the core identification items and the disease phenotype in the real medical scene may be different, it is necessary to establish a real core identification database, and optimize the core identification based on the real core identification database, that is, merge multiple core identification items corresponding to the same disease phenotype to ensure the integrity of the core identification items.

[0119] As an example, assuming that the core identification items of disease phenotype a in the real core identification database are 1, 2, and 3; the core identification items of disease phenotype a in the core identification database are 2, 3, and 4, then the optimized core identification database is 1, 2, 3, and 4. That is, in the embodiment of the present application, the core identification database can be optimized.

[0120] Among them, since some diagnostic terms in the medical record data may not be consistent with the terms in the core identification database, the diagnostic terms in the medical record data can also be standardized.

[0121] Step 2: Determine the target core identification item set based on the real core identification database and the core identification database.

[0122] Specifically, the existing core identification database is optimized according to the real core identification database to obtain an optimized core identification item database; then, the target disease phenotype of the user to be consulted is searched in the optimized core identification item database to determine the core identification item corresponding to the target disease phenotype; then, when the core identification items corresponding to the two disease phenotypes are greater than the preset value, the core identification items corresponding to the two disease phenotypes are classified into the same core identification item set; finally, the target core identification item set is obtained. The target core identification item set can be multiple, and each target core identification item set is a set of similar core identification items represented by different disease phenotypes.

[0123] Furthermore, if Figure 3 As shown, the specific implementation steps of step S203 may include:

[0124] S301: Determine a set of core identification items corresponding to each target disease phenotype.

[0125] It should be understood that there may be multiple target disease phenotypes, and each target disease phenotype may have a corresponding core identification item set, such as the core identification item set of disease phenotype A is (a, b, c, d, e), and the core identification item set of target disease phenotype B is (b, c, d, e). The core identification item set is obtained by matching the target disease phenotype of the target disease of the user to be consulted with the core identification database.

[0126] S302: Based on each set of core identification items, calculate the similarity between each disease phenotype.

[0127] As an example, the core identification items of a disease are (a, b, c, d, e), where a, b, c, d, e are different diseases to be identified; a and β are two disease phenotypes in the disease phenotype classification set, where the core identification items corresponding to disease phenotype a are a, b, c, d, recorded as disease phenotype a (a, b, c, d); the core identification items corresponding to disease phenotype β are a, b, c, e, recorded as disease phenotype β (a, b, c, e). Among them, the diseases to be identified in the core identification items of the disease phenotype that contain a disease are recorded as 1, and the diseases to be identified in the core identification items that do not contain a disease are recorded as 0, so the target vectors of disease phenotype a are (1, 1, 1, 1, 0), and the target vectors of disease phenotype β are (1, 1, 1, 0, 1).

[0128] Furthermore, a weight value may be assigned to the disease to be identified in the target vector corresponding to each of the disease phenotype a and the disease phenotype b, which may be specifically expressed by the following formula (2):

[0129] IMPTNB=log(p(x;=1|y;=1))-log(p(x;=1|y;=0)); (1)

[0130] Wherein, IMPTNB is the logarithmic probability value of the disease x to be identified under the condition that the disease phenotype y appears when the weight value is log(p(x;=1|y;=1)). log(p(x;=1|y;=0)) is the logarithmic probability value of the disease x to be identified under the condition that the disease phenotype y does not appear.

[0131] Specifically, the weight value can be understood as the probability of a certain disease to be identified appearing under the condition of being diagnosed with a certain disease phenotype. For example, there are 10 patients diagnosed with disease phenotype y1, of which 5 patients have diseases to be identified as x1 and x2, and the other 5 patients only have disease to be identified x1, then the probability of disease to be identified x1 appearing in disease phenotype y1 is 100%, that is, the weight value of disease to be identified x1 for disease phenotype y1 is 1, and similarly, the weight value of disease to be identified x2 for disease phenotype y1 is 0.5. It should be understood that each patient may correspond to multiple diseases to be identified, and the weight value of each disease to be identified is related to the number of patients diagnosed with the disease phenotype and the number of patients diagnosed with the disease phenotype who have the disease to be identified.

[0132] In this application, the size of the weight value can represent the importance of each disease to be identified for a certain disease phenotype. The larger the weight value, the more important the corresponding disease to be identified is for the disease phenotype. For example, assuming that the weight values ​​of the diseases to be identified a, b, c, d, and e in the core identification item are all 0.5, then after assigning the weight values, the target vector of disease phenotype a is (1×0.5, 1×0.5, 1×0.5, 1×0.5, 0X0.5), and the target vector of disease phenotype β is (1×0.5, 1×0.5, 1×0.5, 0×0.5, 1X0.5).

[0133] After determining the target vectors of the core identification items corresponding to each disease phenotype, the similarities between the disease phenotypes can be calculated based on the target vectors corresponding to each disease phenotype.

[0134] Specifically, the similarity between disease phenotype a and disease phenotype β can be calculated by the cosine similarity formula, as shown in formula (2):

[0135]

[0136] Wherein, A and B represent two target vectors for calculating similarity. For example, A corresponds to disease phenotype a, and B corresponds to disease phenotype β. When the target vector of disease phenotype a is (0.5, 0.5, 0.5, 0, 0), and the target vector of disease phenotype β is (0.5, 0.5, 0.5, 0, 0.5), the similarity between disease phenotype a and disease phenotype β can be calculated as 0.75 by the above formula (2). Those skilled in the art should know that the method of calculating similarity by the cosine similarity formula is not used to limit the present application, and other similarity calculation methods are also within the scope of protection of the present application, for example, Euclidean distance, Manhattan distance, and Jaccard similarity coefficient.

[0137] S303: Merging the disease phenotypes whose similarity is higher than a preset value to obtain a target core identification item set.

[0138] Specifically, since there may be core identification items with extremely high similarity in the core identification item sets corresponding to each target disease phenotype, in order to further improve the accuracy of the core identification items of the disease phenotype, when calculating the similarity between each disease phenotype, the core identification items corresponding to the disease phenotype with a similarity higher than a preset value can be merged to obtain the target core identification item set.

[0139] S204: Determine the differential diagnosis result of the consulting user according to the target core identification item set.

[0140] It should be understood that after determining the target core identification item set, the differential diagnosis result corresponding to the target disease of the target consulting user can be determined based on the target core identification item set. In this way, the differential diagnosis result can be accurately subdivided into the diseases to be identified associated with the target disease, thereby improving the accuracy of the differential diagnosis result.

[0141] In addition, it can also avoid the deviation of differential diagnosis results caused by errors in the disease diagnosed in the preliminary diagnosis results, thereby improving the accuracy of differential diagnosis results.

[0142] Based on the disease differential diagnosis method provided in the above embodiment, the present application embodiment also provides a disease differential diagnosis device, see Figure 4 , which is a schematic diagram of the structure of the disease differential diagnosis device provided in the embodiment of the present application, combined with Figure 4 As shown, the device 400 provided in the embodiment of the present application may include:

[0143] The acquisition module 401 is used to acquire a disease differential diagnosis database and a disease phenotype classification database; the disease differential diagnosis database includes the corresponding relationship between each disease and each core identification item, and the core identification item includes at least one disease to be identified associated with the corresponding disease; the disease phenotype classification database includes the corresponding relationship between each disease phenotype and each disease;

[0144] The core identification database establishment module 402 is used to establish a core identification database based on the disease differential diagnosis database and the disease phenotype classification database; the core identification database includes the corresponding relationship between each disease phenotype and each core identification item; the corresponding relationship between each disease phenotype and each core identification item is determined by each disease;

[0145] The target core identification item determination module 403 is used to match the target disease phenotype corresponding to the target disease of the consulting user with the core identification database to obtain the target core identification item corresponding to the target disease phenotype;

[0146] The differential diagnosis module 404 is used to determine the differential diagnosis result of the consulting user according to the target core identification item.

[0147] As an example, the differential diagnosis module 404 is used to:

[0148] Determining an actual core identification item corresponding to the target disease; the actual core identification item includes at least one actual disease to be identified associated with the target disease;

[0149] The actual to-be-identified disease in the actual core identification item that is identical to at least one to-be-identified disease in the target core identification item is used as the differential diagnosis result of the inquiring user.

[0150] As an example, when the target disease has a plurality of corresponding target disease phenotypes, the device further includes:

[0151] An acquisition unit, used to acquire target core identification items corresponding to each target disease phenotype;

[0152] a similarity calculation unit, used for calculating the similarity between every two target core identification items in the plurality of target core identification items;

[0153] A merging unit, used for merging two target core identification items corresponding to a target similarity among the multiple similarities to obtain a merged core identification item; the target similarity is a similarity among the multiple similarities that is greater than a preset similarity;

[0154] The differential diagnosis result determination unit is used to use the unmerged target core identification items and the merged core identification items among the multiple target core identification items as the differential diagnosis results of the inquiring user.

[0155] As an example, the similarity calculation unit includes:

[0156] An acquisition subunit is used to acquire an actual core identification item corresponding to the target disease; the actual core identification item includes at least one actual disease to be identified associated with the target disease;

[0157] A determination subunit is used to determine a target vector of each target core identification item according to the actual core identification item; the target vector includes a first element and a second element, the first element includes the disease to be identified in the target core identification item that is the same as the actual disease to be identified in the actual core identification item, and the second element includes the disease to be identified in the target core identification item that is different from the actual disease to be identified in the actual core identification item;

[0158] The calculation subunit is used to calculate the similarity between each two target core identification items according to each target vector.

[0159] As an example, before determining the subunit, the following is also included:

[0160] The standardization processing subunit is used to perform standardization processing on the actual diseases to be identified in the actual core identification items to obtain standard actual core identification items; and to perform standardization processing on each target core identification item to obtain multiple standard target core identification items;

[0161] A determination subunit is used to determine the target vector of each standard target core identification item according to the standard actual core identification item.

[0162] As an example, a disease differential diagnosis database is obtained in the following manner: a first database generation module is used to generate a disease differential diagnosis database according to the correspondence between each disease in a knowledge source and each differential diagnosis; the differential diagnosis includes the corresponding disease to be identified in the diagnosis process.

[0163] As an example, a disease phenotype classification database is obtained in the following way:

[0164] The second database generation module is used for:

[0165] Acquire multiple medical record data; the medical record data includes disease information of corresponding historical consultation users;

[0166] Processing multiple medical record data to obtain multiple standard medical record data;

[0167] Classify multiple standard medical record data based on the pre-trained disease phenotype classification model;

[0168] A disease phenotype classification database is established based on classified multiple standard medical record data.

[0169] The disease differential diagnosis device provided in the embodiment of the present application has the same beneficial effects as the disease differential diagnosis method provided in the above embodiment, so it will not be repeated here.

[0170] The present application also provides a computer device, such as Figure 5 As shown, the computer device 01 is in the form of a general computing device. The components of the computer device 01 may include but are not limited to: one or more processors or processing units 03, a system memory 08, and a bus 04 connecting different system components (including the system memory 08 and the processing unit 03).

[0171] Bus 04 represents one or more of several types of bus structures, including a memory bus or memory controller, a peripheral bus, a graphics acceleration port, a processor or a local bus using any of a variety of bus structures. For example, these architectures include but are not limited to Industry Standard Architecture (ISA) bus, Micro Channel Architecture (MAC) bus, Enhanced ISA bus, Video Electronics Standards Association (VESA) local bus and Peripheral Component Interconnect (PCI) bus.

[0172] The computer device 01 typically includes a variety of computer system readable media. These media can be any available media that can be accessed by the computer device 01, including volatile and non-volatile media, removable and non-removable media.

[0173] The system memory 08 may include computer system readable media in the form of volatile memory, such as random access memory (RAM) 09 and / or cache memory 10. The computer device 01 may further include other removable / non-removable, volatile / non-volatile computer system storage media. By way of example only, the storage system 11 may be used to read and write non-removable, non-volatile magnetic media ( Figure 3 not shown, usually called a "hard drive"). Although Figure 3 Not shown in the figure, a disk drive for reading and writing a removable non-volatile disk (e.g., a "floppy disk"), and an optical disk drive for reading and writing a removable non-volatile optical disk (e.g., a CD-ROM, a DVD-ROM, or other optical media) may be provided. In these cases, each drive may be connected to the bus 04 via one or more data medium interfaces. The memory 08 may include at least one program product having a set (e.g., at least one) of program modules that are configured to perform the functions of the various embodiments of the present invention.

[0174] A program / utility 12 having a set (at least one) of program modules 13 may be stored, for example, in the memory 08, such program modules 13 including but not limited to an operating system, one or more application programs, other program modules, and program data, each of which or some combination thereof may include an implementation of a network environment. The program modules 13 generally perform the functions and / or methods of the embodiments described herein.

[0175] The computer device 01 may also communicate with one or more external devices 02 (e.g., keyboard, pointing device, display 07, etc.), one or more devices that enable a user to interact with the computer device 01, and / or any device that enables the computer device 01 to communicate with one or more other computing devices (e.g., network card, modem, etc.). Such communication may be performed via input / output (I / O) interface 06. Furthermore, the computer device 01 may also communicate with one or more networks (e.g., local area network (LAN), wide area network (WAN) and / or public network, such as the Internet) via network adapter 05. Figure 3 As shown, the network adapter 05 communicates with other modules of the computer device 01 via the bus 04. It should be understood that although Figure 3 Not shown, other hardware and / or software modules may be used in conjunction with computer 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.

[0176] The processor unit 03 executes various functional applications and data processing by running the programs stored in the system memory 08, such as implementing the disease differential diagnosis method provided in any embodiment of the present application.

[0177] The embodiments of the present application also provide a computer storage medium for implementing the solutions provided by the embodiments of the present application.

[0178] The computer storage medium stores codes, and when the codes are executed, the device executing the codes implements the disease differential diagnosis method described in any embodiment of the present application.

[0179] It should be noted that each embodiment in this specification is described in a progressive manner, and the same or similar parts between the embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the device and equipment embodiments, since they are basically similar to the method embodiments, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiments. The device and equipment embodiments described above are merely schematic, wherein the units described as separate components may or may not be physically separated, and the components indicated as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. A person of ordinary skill in the art can understand and implement it without paying creative labor.

[0180] The "first" and "second" in the names such as "first" and "second" (if any) mentioned in the embodiments of the present application are only used as name identifiers and do not represent the first or second in order.

[0181] Through the description of the above implementation methods, it can be known that those skilled in the art can clearly understand that all or part of the steps in the above-mentioned embodiment method can be implemented by means of software plus a general hardware platform. Based on such an understanding, the technical solution of the present application can be embodied in the form of a software product, which can be stored in a storage medium, such as a read-only memory (ROM) / RAM, a magnetic disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network communication device such as a router) to execute the methods described in each embodiment of the present application or some parts of the embodiments.

[0182] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any changes or substitutions that can be easily thought of by a person skilled in the art within the technical scope disclosed in the present application should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. A method for differential diagnosis of a disease, characterized in that: The method comprises: Acquire a disease differential diagnosis database and a disease phenotype classification database; the disease differential diagnosis database includes a correspondence between each disease and each core identification item, the core identification item includes at least one disease to be identified associated with the corresponding disease; the disease phenotype classification database includes a correspondence between each disease phenotype and each disease; A core identification database is established based on the disease differential diagnosis database and the disease phenotype classification database; the core identification database includes the corresponding relationship between each disease phenotype and each core identification item; the corresponding relationship between each disease phenotype and each core identification item is determined by each disease; Matching the target disease phenotype corresponding to the target disease of the consulting user with the core identification database to obtain the target core identification item corresponding to the target disease phenotype; The differential diagnosis result of the consulting user is determined according to the target core identification item.

2. The method according to claim 1, characterized in that The step of determining the differential diagnosis result of the consulting user according to the target core identification item includes: Determining an actual core identification item corresponding to the target disease; the actual core identification item includes at least one actual disease to be identified associated with the target disease; The actual to-be-identified disease in the actual core identification item that is identical to at least one to-be-identified disease in the target core identification item is used as the differential diagnosis result of the consulting user.

3. The method according to claim 1, characterized in that: When the target disease has a plurality of corresponding target disease phenotypes, the method further comprises: Obtain target core identification items corresponding to each target disease phenotype; Calculating the similarity between every two target core identification items in the plurality of target core identification items; Merging two target core identification items corresponding to a target similarity among the multiple similarities to obtain a merged core identification item; the target similarity is a similarity among the multiple similarities that is greater than a preset similarity; The unmerged target core identification items and the merged core identification items among the multiple target core identification items are used as the differential diagnosis results of the consulting user.

4. The method according to claim 3, characterized in that The calculating the similarity between every two target core identification items in the plurality of target core identification items comprises: Acquire an actual core identification item corresponding to the target disease; the actual core identification item includes at least one actual disease to be identified associated with the target disease; Determine a target vector for each target core identification item according to the actual core identification item; the target vector includes a first element and a second element, the first element includes the disease to be identified in the target core identification item that is the same as the actual disease to be identified in the actual core identification item, and the second element includes the disease to be identified in the target core identification item that is different from the actual disease to be identified in the actual core identification item; The similarity between each two target core identification items is calculated according to the target vectors.

5. The method according to claim 4, characterized in that Before determining the target vector of each target core identification item according to the actual core identification item, the method further comprises: Performing standardization processing on the actual diseases to be identified in the actual core identification items to obtain standard actual core identification items; and performing the standardization processing on each target core identification item to obtain multiple standard target core identification items; The determining the target vector of each target core identification item according to the actual core identification item includes: determining the target vector of each standard target core identification item according to the standard actual core identification item.

6. The method according to claims 1-5, characterized in that: The disease differential diagnosis database is obtained in the following manner: generating the disease differential diagnosis database according to the correspondence between each disease in the knowledge source and each differential diagnosis; the differential diagnosis includes the diseases to be differentiated in the diagnosis process of the corresponding disease.

7. The method according to claims 1-5, characterized in that: The disease phenotype classification database is obtained by the following method: Acquire multiple medical record data; the medical record data includes disease information of corresponding historical consultation users; Processing the plurality of medical record data to obtain a plurality of standard medical record data; Classifying the plurality of standard medical record data according to a pre-trained disease phenotype classification model; The disease phenotype classification database is established based on the classified plurality of standard medical record data.

8. A disease differential diagnosis device, characterized in that: The device comprises: An acquisition module is used to acquire a disease differential diagnosis database and a disease phenotype classification database; the disease differential diagnosis database includes a correspondence between each disease and each core identification item, and the core identification item includes at least one disease to be identified associated with the corresponding disease; the disease phenotype classification database includes a correspondence between each disease phenotype and each disease; A core identification database establishment module is used to establish a core identification database according to the disease differential diagnosis database and the disease phenotype classification database; the core identification database includes the corresponding relationship between each disease phenotype and each core identification item; the corresponding relationship between each disease phenotype and each core identification item is determined by each disease; A target core identification item determination module is used to match the target disease phenotype corresponding to the target disease of the consulting user with the core identification database to obtain the target core identification item corresponding to the target disease phenotype; The differential diagnosis module is used to determine the differential diagnosis result of the consulting user according to the target core identification item.

9. A computer device, characterized in that: include: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the disease differential diagnosis method according to any one of claims 1 to 7 is implemented.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores instructions, and when the instructions are executed on a terminal device, the terminal device executes the disease differential diagnosis method according to any one of claims 1 to 7.

Images