Data processing method, device and computing equipment

Abstract

The embodiment of the present application provides a data processing method, device and computing equipment. The method includes: obtaining medical data of a target user; based on the medical data, using a classification prediction model to predict the target diagnostic group category to which the target user belongs; wherein the classification prediction model is pre-trained using the medical sample data corresponding to the sample user and the diagnostic group category to which it belongs. The technical solution provided by the embodiment of the present application improves the efficiency and accuracy of disease diagnosis grouping.

Description

Technical Field

[0001] The embodiments of the present application relate to the field of computer application technology, and in particular, to a data processing method, apparatus and computing device. Background Art

[0002] DRGs (Diagnosis Related Groups) is a disease diagnosis classification mechanism that is widely used in the medical field. It is used to classify and group patients. Medical compensation can be achieved based on the divided diagnostic grouping categories.

[0003] At present, DRGs are implemented by professionals who group patients according to the DRGS standards based on the data on the first page of the medical record filled out by the patient after discharge. However, due to human factors and other influences, efficient and accurate disease diagnosis grouping cannot be achieved. Summary of the invention

[0004] The embodiments of the present application provide a data processing method, apparatus and computing device to solve the technical problem that efficient and accurate disease diagnosis grouping cannot be achieved in the prior art.

[0005] In a first aspect, an embodiment of the present application provides a medical data processing method, comprising:

[0006] Obtain medical data of target patients;

[0007] Based on the medical data, using a classification prediction model to predict the target diagnostic group category to which the target patient belongs;

[0008] The classification prediction model is pre-trained using the medical sample data corresponding to the sample patients and the diagnostic grouping categories to which they belong.

[0009] In a second aspect, an embodiment of the present application provides a medical data processing method, comprising:

[0010] Determine the medical sample data of the sample patients and the diagnostic group category to which they belong;

[0011] The classification prediction model is trained using the medical sample data and the diagnostic grouping categories.

[0012] In a third aspect, an embodiment of the present application provides a medical data processing method, including:

[0013] Obtain medical data of target patients;

[0014] Based on the medical data, using a classification prediction model to predict the target diagnostic group category to which the target patient belongs;

[0015] Execute corresponding task processing operations according to the target diagnosis group category;

[0016] The classification prediction model is pre-trained using the medical sample data corresponding to the sample patients and the diagnostic grouping categories to which they belong.

[0017] In a fourth aspect, an embodiment of the present application provides a data processing method, including:

[0018] Obtain medical data of target users;

[0019] Based on the medical data, using a classification prediction model to predict the target diagnosis group category to which the target user belongs;

[0020] The classification prediction model is pre-trained using the medical sample data corresponding to the sample users and the diagnostic grouping categories to which they belong.

[0021] In a fifth aspect, an embodiment of the present application provides a data processing method, including:

[0022] Determine the medical sample data of the sample user and the diagnostic group category to which it belongs;

[0023] The classification prediction model is trained using the medical sample data and the diagnostic grouping categories.

[0024] In a sixth aspect, an embodiment of the present application provides a data processing method, including:

[0025] Obtain medical data of target users;

[0026] Based on the medical data, using a classification prediction model to predict the target diagnosis group category to which the target user belongs;

[0027] Execute corresponding task processing operations according to the target diagnosis group category;

[0028] The classification prediction model is pre-trained using the medical sample data corresponding to the sample users and the diagnostic grouping categories to which they belong.

[0029] In a seventh aspect, an embodiment of the present application provides a data processing device, including:

[0030] A first acquisition module is used to acquire medical data of a target user;

[0031] A category prediction module, used to predict the target diagnosis group category to which the target user belongs based on the medical data using a classification prediction model;

[0032] The classification prediction model is pre-trained using the medical sample data corresponding to the sample users and the diagnostic grouping categories to which they belong.

[0033] In an eighth aspect, an embodiment of the present application provides a data processing device, including:

[0034] The second acquisition module is used to determine the medical sample data of the sample user and the diagnostic group category to which it belongs;

[0035] The model training module is used to train a classification prediction model using the medical sample data and the diagnostic grouping categories.

[0036] In a ninth aspect, an embodiment of the present application provides a computing device, including a processing component and a storage component;

[0037] The storage component stores one or more computer instructions; the one or more computer instructions are used to be called and executed by the processing component to implement the data processing method described in the third aspect above.

[0038] In a tenth aspect, an embodiment of the present application provides a computing device, including a processing component and a storage component;

[0039] The storage component stores one or more computer instructions; the one or more computer instructions are used to be called and executed by the processing component to implement the data processing method described in the fourth aspect above.

[0040] In the eleventh aspect, a computer storage medium is provided in an embodiment of the present application, storing a computer program, and when the computer program is executed by a computer, it implements the data processing method described in the third aspect above.

[0041] In the twelfth aspect, an embodiment of the present application provides a computer storage medium storing a computer program, which, when executed by a computer, implements the data processing method as described in the fourth aspect above.

[0042] In an embodiment of the present application, a classification prediction model is pre-trained based on the medical sample data corresponding to the sample user and the diagnostic group category to which it belongs. Therefore, for a target user, after obtaining the target user's medical data, the classification prediction model is input to predict the diagnostic group category to which the target user belongs, thereby improving the efficiency and accuracy of disease diagnosis grouping.

[0043] These and other aspects of the present application will become more clearly understood in the description of the following embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, a brief introduction will be given below to the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0045] Figure 1 A flow chart showing an embodiment of a data processing method provided by the present application is shown;

[0046] Figure 2 A flowchart showing another embodiment of a data processing method provided by the present application is shown;

[0047] Figure 3 A flowchart showing another embodiment of a data processing method provided by the present application is shown;

[0048] Figure 4 A schematic diagram of feature extraction in a practical application of an embodiment of the present application is shown;

[0049] Figure 5 A schematic diagram of model training and application in a practical application of an embodiment of the present application is shown;

[0050] Figure 6 A schematic diagram showing the structure of an embodiment of a data processing device provided by the present application is shown;

[0051] Figure 7 A schematic diagram showing the structure of an embodiment of a computing device provided by the present application is shown;

[0052] Figure 8 A structural schematic diagram of another embodiment of a data processing device provided by the present application is shown;

[0053] Fig. 9 A structural schematic diagram of another embodiment of a computing device provided by the present application is shown. DETAILED DESCRIPTION

[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.

[0055] In some of the processes described in the specification and claims of this application and the above-mentioned figures, multiple operations that appear in a specific order are included, but it should be clearly understood that these operations may not be executed in the order in which they appear in this article or executed in parallel. The serial numbers of the operations, such as 101, 102, etc., are only used to distinguish between different operations, and the serial numbers themselves do not represent any execution order. In addition, these processes may include more or fewer operations, and these operations may be executed in sequence or in parallel. It should be noted that the descriptions of "first", "second", etc. in this article are used to distinguish different messages, devices, modules, etc., do not represent the order of precedence, and do not limit the "first" and "second" to be different types.

[0056] The technical solution of the present application is mainly used in scenarios where patients are grouped according to disease diagnoses (hereinafter referred to as diagnostic groups) in the medical field, for example, hospitalized patients are grouped according to the DRGs (Diagnosis Related Groups) standards.

[0057] Taking DRGs grouping as an example, it is currently usually done by professionals according to different DRGs standards based on the data on the first page of the medical record filled out by inpatients when they are discharged. This is inefficient and has poor accuracy. In addition, since different institutions may use different DRGS standards, the grouping method is not universal.

[0058] In order to improve the efficiency and accuracy of diagnostic grouping, the inventors have proposed the technical solution of the present application after a series of studies. In the embodiments of the present application, a classification prediction model is pre-trained based on the medical sample data corresponding to the sample user and the diagnostic grouping category to which they belong. Therefore, for a target user, after obtaining the target user's medical data, the classification prediction model can be input to predict the diagnostic grouping category to which the target user belongs, thereby improving the efficiency and accuracy of diagnostic grouping. Moreover, diagnostic grouping can be achieved without understanding the diagnostic grouping standards. The grouping method is fast, efficient, and highly versatile.

[0059] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions 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 those skilled in the art without creative work are within the scope of protection of this application.

[0060] Figure 1 A flowchart of an embodiment of a data processing method provided in an embodiment of the present application, the method may include the following steps:

[0061] 101: Obtain the target user’s medical data.

[0062] As an optional implementation method, obtaining the medical data of the target user can be: obtaining the medical record homepage data from the medical record homepage corresponding to the target user. The medical record homepage is a data summary formed by medical staff in medical institutions using text, symbols, codes, numbers, etc. to summarize the relevant information of the patient during hospitalization in a specific form after the patient is discharged from the hospital. According to the regulatory requirements of different countries, the medical record homepage mainly records the personal attribute information of the hospitalized patient, the hospitalization diagnosis (including the main diagnosis, secondary diagnosis, etc.), the hospitalization medical expenses and other information.

[0063] Therefore, the target user's medical record homepage data can be obtained from the medical record homepage, including the personal attribute information, hospitalization diagnosis, hospitalization medical expenses, etc. described above. The medical record homepage data is usually formatted data that meets national standards, such as standardized data types and data ranges.

[0064] Since the data on the medical record homepage is generated only after the patient is discharged, the target user may refer to discharged patients. Therefore, diagnostic grouping based on the data on the medical record homepage can realize the diagnostic grouping category of discharged patients. For example, DRGs is the grouping of discharged patients, so that corresponding processing can be carried out based on the DRGs grouping category to which the discharged patients belong, such as medical insurance expense compensation, medical institution performance evaluation, etc.

[0065] However, it is impossible to diagnose and group users who are in the process of diagnosis and treatment based on the data on the medical record homepage. In actual applications, there is a need to diagnose and group users who are still in the process of diagnosis and treatment. Therefore, as another optional implementation method, obtaining the medical data of the target user can be: obtaining the medical record data produced by the target user during the diagnosis and treatment process.

[0066] In practical applications, the target user refers to the target patient to be classified, which may be an inpatient or an outpatient.

[0067] Medical record data may refer to the records of medical personnel's medical activities such as examination, diagnosis, and treatment of patients' occurrence, development, and outcome. For example, it may include: admission diagnosis, surgical operation records, symptom records, medical orders, nursing records, death records, user personal attribute information such as age, gender, etc. Medical record data is recorded by medical personnel during the medical activity process and is usually unformatted data, such as text or images.

[0068] The medical record data produced by the target users during the diagnosis and treatment process can be obtained from the medical data system. With the advent of the intelligent era and the data era, medical institutions will store and manage the diagnosis and treatment conditions of different patients in electronic form to facilitate data search and management. The medical data system can deploy computer systems that can process data for medical institutions.

[0069] Optionally, obtaining the medical record data of the target user generated during the diagnosis and treatment process may include: detecting that the medical data system updates the medical record of the target user during the diagnosis and treatment process of the target user, and obtaining the medical record data of the target user from the medical data system.

[0070] That is, as the diagnosis and treatment process is carried out, when medical staff updates the medical records of the target user, it can trigger the acquisition of the target user's medical record data, so as to realize dynamic and real-time diagnosis grouping prediction for the target user.

[0071] 102: Based on medical data, use the classification prediction model to predict the target diagnosis group category to which the target user belongs.

[0072] Among them, the classification prediction model can be pre-trained using the medical sample data corresponding to the sample users and the diagnostic group categories to which they belong.

[0073] Among them, the model training process of the classification pre-model can be found in Figure 2 The data processing method shown may include the following steps:

[0074] 201: Determine the medical sample data of the sample user and the diagnostic group category to which it belongs.

[0075] The sample users may refer to sample patients, specifically historical patients who have been classified into diagnostic grouping categories in the medical institution where the target user is located, so that the medical sample data of the sample users and their diagnostic grouping categories may be obtained from the historical medical records.

[0076] As an optional method, the medical sample data may refer to the medical record front page data corresponding to the sample user; thereby, the classification prediction model may make diagnosis grouping predictions for the target user based on the number of the target user's medical record front pages.

[0077] As another optional method, the medical sample data may refer to the medical record data generated by the sample user during the diagnosis and treatment process, so that the classification prediction module can make a diagnosis group prediction for the target user based on the medical record data generated by the target user during the diagnosis and treatment process.

[0078] 202: Use medical sample data and diagnostic grouping categories to train classification prediction models.

[0079] In an embodiment of the present application, a classification prediction model can be pre-trained based on the medical sample data corresponding to the sample users and the diagnostic group categories to which they belong. Therefore, for a target user, after obtaining the target user's medical data, the classification prediction model can be input to predict the diagnostic group category to which the target user belongs, thereby improving grouping efficiency and accuracy.

[0080] In practical applications, various task processing operations can be performed based on the target diagnosis group category to which the target user belongs. Figure 3 In the data processing method shown, the method may include the following steps:

[0081] 301: Obtain the medical data of the target user.

[0082] 302: Based on the medical data, a classification prediction model is used to predict the target diagnosis group category to which the target user belongs.

[0083] 304: Execute corresponding task processing operations according to the target diagnosis group category.

[0084] In an optional implementation, according to the target diagnosis group category, executing the corresponding task processing operation may include:

[0085] Based on the target diagnostic grouping category, the target user's medical institution will be evaluated for services or compensated for costs.

[0086] For example, different expense compensation standards corresponding to different diagnostic grouping categories can be preset, and medical institutions can be compensated for expenses according to the expense compensation standards corresponding to the target diagnostic grouping categories. In particular, when the target user is a user participating in medical insurance, the insurance institution determines the target diagnostic grouping category of the target user according to the technical solution of this application, thereby compensating the medical institution where the target user is located.

[0087] For example, different diagnostic grouping categories can be pre-set to correspond to different disease difficulty coefficients. Then, combined with the disease difficulty coefficient corresponding to the target diagnostic grouping category, a service evaluation of the medical institution where the target user is located can be conducted, such as performance evaluation. The higher the disease difficulty coefficient corresponding to the target diagnostic grouping category, the stronger the disease treatment ability of the medical institution.

[0088] In another optional implementation, according to the target diagnosis group category, executing the corresponding task processing operation may include:

[0089] Generate diagnosis and treatment recommendation information for target users based on the target diagnosis grouping category.

[0090] As can be seen from the above description, the medical data of the target user can be the medical record data generated by the target user during the diagnosis and treatment process, so that the classification prediction model can realize the dynamic diagnosis grouping of the target user during the diagnosis and treatment process. Therefore, based on the target diagnosis grouping category and the medical attribute information related to the target user, the diagnosis and treatment recommendation information can be obtained, so that the medical institution can determine whether to adjust the diagnosis and treatment of the target user in combination with the diagnosis and treatment recommendation information.

[0091] Optionally, the target diagnosis group category and the medical attribute information related to the target user may be combined to generate diagnosis and treatment recommendation information for the target user.

[0092] The diagnosis and treatment recommendation information can be used to prompt whether to adjust the medical plan for the target user, etc.

[0093] The medical attribute information related to the target user may include, for example, the actual diagnosis result, the target department and / or the current consumption information of the target user.

[0094] For example, based on the target diagnosis group category and the medical attribute information related to the target user, generating diagnosis and treatment recommendation information for the target user may include:

[0095] If the diagnosis result corresponding to the target diagnosis group category is different from the actual diagnosis result, diagnosis and treatment recommendation information for the target user is generated.

[0096] The actual diagnosis result may refer to the main diagnosis recorded in the target user's medical data, and different diagnostic grouping categories may pre-set corresponding diagnostic results. If the diagnostic result corresponding to the target diagnostic grouping category is different from the actual diagnostic result, indicating that there may be problems with the diagnosis and treatment of the target user, diagnosis and treatment recommendation information may be generated to prompt the medical institution whether to adjust the medical plan for the target user, etc.

[0097] For another example, based on the target diagnosis group category and the medical attribute information related to the target user, generating diagnosis and treatment suggestion information for the target user may include:

[0098] If the target diagnostic grouping category is different from the diagnostic grouping category corresponding to the target department where the target user is located, diagnostic and treatment recommendation information is generated for the target user.

[0099] In actual applications, in order to improve the treatment capabilities of certain medical departments or to standardize the disease treatment scope of different medical departments so that patients can receive targeted treatment, the diagnostic grouping category corresponding to the target department can usually be pre-set according to the disease difficulty coefficient corresponding to the diagnostic grouping category. Therefore, if the target diagnostic grouping category is different from the diagnostic grouping category corresponding to the target department where the target user is located, it indicates that the target department is treating patients who are no longer within the treatment scope. Therefore, diagnosis and treatment recommendation information can be generated to prompt the medical institution whether to adjust the medical plan for the target user, such as changing the treatment department.

[0100] For another example, based on the target diagnosis group category and the medical attribute information related to the target user, generating diagnosis and treatment suggestion information for the target user may include:

[0101] If the consumption cost corresponding to the target diagnostic group category is inconsistent with the current consumption cost of the target user, diagnosis and treatment recommendation information for the target user is generated.

[0102] The current consumption information includes the current consumption fee, which may refer to the medical treatment fee currently spent.

[0103] For example, if the consumption cost corresponding to the target diagnostic group category is lower than the current consumption cost of the target user, it indicates that the diagnosis and treatment charges for the target user are unreasonable, then diagnosis and treatment recommendation information can be generated to prompt the user whether to adjust the medical plan, etc.

[0104] Optionally, treatment recommendation information may be generated by combining one or more judgment results, including that the diagnosis result corresponding to the target diagnosis grouping category is different from the actual diagnosis result, the target diagnosis grouping category is different from the diagnosis grouping category corresponding to the target department where the target user is located, and the consumption cost corresponding to the target diagnosis grouping category is inconsistent with the current consumption cost of the target user.

[0105] It should be noted that the above is only an example of combining the target diagnosis group category to generate diagnosis and treatment recommendation information for the target user. By adopting the technical solution of this application example, it is possible to dynamically predict the diagnosis group of the target user during the diagnosis and treatment of the target user, so that the medical plan for the target user can be dynamically adjusted in combination with the diagnosis group result. In the DRGs application scenario, since DRGs grouping is mainly used for medical insurance compensation and performance evaluation of medical institutions, by dynamically predicting the DRGs grouping category of the target user during the diagnosis and treatment process, medical insurance compensation and performance evaluation can be predicted, and the medical plan can be adjusted in combination with the prediction results.

[0106] Of course, other task processing operations can also be performed based on the target diagnostic group category. This application does not specifically limit this. The technical solution of the embodiment of this application is used to predict the target diagnostic group category of the target user, and any task processing based on the target diagnostic group should be within the scope of protection of this application.

[0107] As can be seen from the foregoing description, medical data may include various types of medical data, such as personal attribute data, doctor's order data, surgical operation record data, symptom record data, etc. In order to improve the accuracy of data processing, in some embodiments, based on medical data, using a classification prediction model to identify the target diagnosis group category to which the target user belongs may include:

[0108] extracting feature data of at least one feature type from the medical data;

[0109] Based on the feature data of at least one feature type, a classification prediction model is used to identify the target diagnosis group category to which the target user belongs.

[0110] Optionally, feature data of at least one feature type may be extracted from data of at least one medical record type.

[0111] The classification of feature types may be the same as or different from the classification of medical record types.

[0112] In an actual application, the classification of feature types may be different from the classification of medical record types. For example, feature types may include gender and age, which are both personal attribute information in the medical record type. Therefore, feature data of different feature types can be extracted from data of the same medical record type, and feature data of the same feature type can also be extracted from data of different medical record types.

[0113] Similarly, when training a classification prediction model, in some embodiments, using medical sample data and diagnostic grouping categories, training the classification prediction model may include:

[0114] Extracting feature sample data of at least one feature type from the medical sample data;

[0115] The classification prediction model is trained using feature sample data of at least one feature type and diagnostic grouping categories.

[0116] Among them, the division of feature types can be carried out in combination with actual conditions, and this application does not make specific restrictions on this. In actual applications, the key feature types for diagnostic grouping can at least include diagnostic category features and surgical operation category features. The diagnostic category feature is used to characterize the disease diagnosis information of the target user, and the surgical operation category feature is used to characterize the surgical operation information of the target user, and the disease diagnosis information and surgical operation information are more important for diagnostic grouping.

[0117] Therefore, in some embodiments, extracting feature data of at least one feature type from the medical data may include:

[0118] Extracting a first number of diagnosis classification codes whose similarity to the medical data meets a first similarity requirement from a diagnosis classification code dictionary, and extracting a second number of surgical operation classification codes whose similarity to the medical data meets a second similarity requirement from a surgical operation classification code dictionary;

[0119] Obtaining diagnostic category characteristics based on a first number of diagnostic classification codes;

[0120] A surgical procedure category feature is obtained based on a second number of diagnosis classification codes.

[0121] Among them, the diagnosis classification code dictionary includes diagnosis classification codes corresponding to different diagnosis names; the surgery classification code dictionary includes surgery operation classification codes corresponding to different surgery operation names.

[0122] Among them, the diagnosis classification code can adopt the diagnosis ICD (International Classification of Diseases) code, and the surgical operation classification code can adopt the surgical operation ICD code.

[0123] For ease of understanding, the following are the diagnostic classification codes corresponding to several diagnostic names:

[0124] Diagnosis name: Infectious gastroenteritis or colitis--------Diagnosis classification code: L1-1A0;

[0125] Diagnosis name: Bacterial intestinal infection-----------------Diagnosis classification code: L2-1A0, etc.

[0126] The following are the surgical operation classification codes corresponding to several surgical operation names:

[0127] Surgical procedure name: Therapeutic ultrasound of head and neck vasculature------Surgical procedure classification code: 00.0101;

[0128] Name of surgical procedure: Cardiovascular therapeutic ultrasound---------Surgical procedure classification code: 00.0201, etc.

[0129] Among them, when the medical data is the medical record homepage data, the medical record homepage data includes the diagnosis classification code and the surgical operation classification code of the target user. Therefore, the diagnosis category feature can be directly obtained based on the diagnosis classification code in the medical record homepage data; the surgical operation category feature can be obtained based on the surgical operation classification code in the medical record homepage data.

[0130] When the medical data is the medical record data generated during the diagnosis and treatment of the target user, the medical data is recorded by medical staff and is usually not standardized like the data on the front page of the medical record. Therefore, it is necessary to perform feature analysis on the medical data, and dictionary matching can be used to determine the corresponding feature data.

[0131] The step of extracting a first number of diagnosis classification codes from the diagnosis classification code dictionary whose similarity to the medical data meets a first similarity requirement, and extracting a second number of surgical operation classification codes from the surgical operation classification code dictionary whose similarity to the medical data meets a second similarity requirement may include:

[0132] Acquire at least one first medical record text related to the diagnosis classification in the medical data;

[0133] Acquire at least one second medical record text related to the surgical operation classification in the medical data;

[0134] Extracting at least one first keyword related to the diagnosis classification from the first medical record text;

[0135] Extracting at least one second keyword related to the surgical operation classification from the second medical record text;

[0136] Performing similarity matching between the first keyword and the diagnosis classification code dictionary to obtain a diagnosis classification code that meets a first similarity requirement with the first keyword;

[0137] The second keyword is matched with the surgical operation classification code dictionary for similarity to obtain a surgical operation classification code that meets the first similarity requirement with the second keyword.

[0138] Since medical data may include data of multiple medical record types, such as surgical operation records, medical orders, diagnosis records, etc., which usually exist in text form, medical record texts of different medical record types can be combined to first obtain at least one first medical record text related to the diagnosis classification and at least one second medical record text related to the surgical operation classification.

[0139] For each first medical record text, at least one first keyword related to the diagnosis classification can be extracted; and for each second medical record text, at least one second keyword related to the surgical operation classification can be extracted.

[0140] Among them, there are many ways to implement keyword extraction. For example, the first keyword extraction can be implemented using a diagnosis classification keyword extraction model, and the second keyword extraction can be implemented using a surgical operation classification keyword extraction model. The diagnosis classification keyword extraction model can be pre-trained based on sample text and diagnosis classification sample keywords annotated in the sample text, and the surgical operation classification keyword extraction model can be pre-trained based on sample text and surgical operation classification sample keywords annotated in the sample text. Of course, model training can also be implemented using an unsupervised training method, and this application does not make specific limitations on this.

[0141] Wherein, for each first keyword, similarity matching can be performed with the diagnosis classification code dictionary to obtain a diagnosis classification code that meets the first similarity requirement with each first keyword, so that a first number of diagnosis classification codes can be obtained correspondingly based on at least one first keyword;

[0142] For each second keyword, similarity matching can be performed with the surgical operation classification code dictionary to obtain a surgical operation classification code that meets the second similarity requirement for each second keyword, so that at least one second keyword can correspond to a second number of surgical operation classification codes.

[0143] The first similarity requirement may, for example, refer to a similarity greater than a first similarity threshold, and the second similarity requirement may, for example, refer to a similarity greater than a second similarity threshold, and so on.

[0144] Among them, the similarity matching of the first keyword with the diagnosis classification coding dictionary can specifically refer to matching with the diagnosis classification name in the diagnosis classification coding dictionary, so as to obtain the diagnosis classification code corresponding to the diagnosis classification name that meets the first similarity requirement with the first keyword; the similarity matching of the second keyword with the surgical operation classification coding dictionary can specifically refer to matching with the surgical operation classification name in the surgical operation classification coding dictionary, so as to obtain the surgical operation classification code corresponding to the surgical operation classification name that meets the second similarity requirement with the second keyword.

[0145] Among them, the similarity matching between words can be achieved by calculating the cosine similarity or Euclidean distance between the two words. Of course, the two words can be first converted into word vectors and then the similarity is calculated. This application does not specifically limit the method of calculating the similarity.

[0146] For ease of understanding, Figure 4 In the feature extraction diagram shown, at least one first medical record text 402 related to the diagnosis classification and at least one second medical record text 403 related to the surgical operation classification can be determined from the medical data 401; the first medical record text can be subjected to NLP (Natural Language Processing), word segmentation and extraction of the first keyword 404, and the second medical record text can be subjected to NLP processing, word segmentation and extraction of the second keyword 405; each first keyword is matched with the diagnosis classification code dictionary 406 for similarity, and finally a first number of diagnosis classification codes 407 can be obtained; each second keyword is matched with the surgical operation classification code dictionary 408 for similarity, and finally a second number of surgical operation classification codes 409 can be obtained; based on the first number of diagnosis classification codes, a diagnosis category feature 410 is obtained; based on the second number of diagnosis classification codes, a surgical operation category feature 411 is obtained.

[0147] As an optional implementation, obtaining the diagnosis category feature based on the first number of diagnosis category codes may include: concatenating the first number of diagnosis category codes in order of priority to obtain a first text; converting the first text into a text vector to obtain the diagnosis category feature;

[0148] Obtaining surgical operation category features based on the second number of diagnosis classification codes may include: concatenating the second number of surgical operation classification codes in order of priority to obtain a second text; and converting the second text into a text vector to obtain the surgical operation category features.

[0149] The priority order of the first number of diagnosis classification codes may be determined according to the degree of similarity with the medical data. The higher the similarity, the higher the priority order. Of course, the priority order of different diagnosis classification codes may also be preset.

[0150] The priority order of the second number of surgical operation classification codes may be determined according to the similarity with the medical data. The higher the similarity, the higher the priority order. Of course, the priority order of different surgical operation classification codes may also be preset.

[0151] Among them, converting the first text into a text vector can be converting each diagnostic classification code in the first text into a corresponding word vector, and converting the second text into a text vector can be converting each surgical operation classification code in the second text into a corresponding word vector. Therefore, converting the first text into a text vector and converting the second text into a text vector can be achieved by using word2vec (word to vector, a word vector conversion technology) or fasttext (a word vector conversion technology) and other word vector models.

[0152] As another optional implementation, obtaining the diagnosis category feature based on the first number of diagnosis category codes may include: converting the first number of diagnosis category codes into first feature vectors respectively, and fusing the first number of first feature vectors to obtain the diagnosis category feature;

[0153] Then obtaining surgical operation category features based on the second number of surgical operation classification codes may include:

[0154] The second number of surgical operation classification codes are converted into second feature vectors respectively, and the second number of second feature vectors are fused to obtain surgical operation category features.

[0155] Among them, converting the diagnosis classification code into the first feature vector and converting the surgical operation classification code into the second feature vector can be implemented using word2vec, fasttext and other word vector models, and this application does not make specific limitations on this.

[0156] One way to fuse the first number of first feature vectors to obtain the diagnosis category feature is to concatenate the first number of first feature vectors to obtain the diagnosis category feature, that is, the diagnosis category feature can be obtained by directly concatenating them together.

[0157] In another manner, fusing a first number of first eigenvectors may be weighted fusion of the first number of first eigenvectors, for example, weighted summation is performed when the weight coefficient is less than 1. Fusing a second number of second eigenvectors may be weighted fusion of the second number of second eigenvectors, etc., wherein the weight coefficient corresponding to each first eigenvector may refer to the weight coefficient of the corresponding diagnosis classification code, and the weight coefficients of different diagnosis classification codes may be pre-set, and of course may also be determined in combination with the priority order of the diagnosis classification codes; the weight coefficient corresponding to each second eigenvector may be the weight coefficient of the corresponding surgical operation classification code, and the weight coefficients of different surgical operation classification codes may be pre-set, and may also be determined in combination with the priority order of the surgical operation classification codes.

[0158] Therefore, in some embodiments, fusing the first number of first feature vectors to obtain the diagnosis category feature may include:

[0159] Determine the weight coefficients corresponding to the first number of first feature vectors according to the priority order of the first number of diagnosis classification codes; perform weighted fusion of the first number of first feature vectors according to the weight coefficients corresponding to the first number of first feature vectors to obtain diagnosis category features;

[0160] Then fusing the second number of second feature vectors to obtain the surgical operation category features may include: determining the weight coefficients corresponding to the second number of second feature vectors according to the priority order of the second number of surgical operation classification codes; and weightedly fusing the second number of second feature vectors according to the weight coefficients corresponding to the second number of second feature vectors to obtain the surgical operation category features.

[0161] For example, assuming that there are P diagnostic classification codes, the corresponding P first eigenvectors can represent V i , the weight coefficient corresponding to each diagnostic classification code can be expressed as W i , i = 1, 2, 3...p, then the diagnostic category feature It can be expressed as:

[0162]

[0163] Assuming that there are q surgical operation classification codes, the corresponding q second feature vectors can be expressed as X i , the weight coefficient corresponding to each diagnostic classification code is Y i , i = 1, 2, 3...p, then the diagnostic category feature It can be expressed as:

[0164]

[0165] Furthermore, in some embodiments, extracting feature data of at least one feature type from the medical data may include:

[0166] Extracting a first number of diagnosis classification codes whose similarity to the medical data meets a first similarity requirement from a diagnosis classification code dictionary, and extracting a second number of surgical operation classification codes whose similarity to the medical data meets a second similarity requirement from a surgical operation classification code dictionary;

[0167] splicing the first number of diagnosis classification codes and the second number of surgical operation classification codes to form a third text;

[0168] The third text is converted into a text vector to obtain feature data.

[0169] That is, the first number of diagnosis classification codes and the second number of surgical operation classification codes may be collected together to form a third text, and then uniformly converted into text vectors to obtain a feature data.

[0170] Among them, converting the third text into a text vector may be converting each code in the third text into a corresponding word vector, which may be implemented by using a word vector model such as word2vec or fasttext.

[0171] Among them, the specific implementation methods of extracting a first number of diagnostic classification codes whose similarity with the medical data meets the first similarity requirement from the diagnostic classification coding dictionary, and extracting a second number of surgical operation classification codes whose similarity with the medical data meets the second similarity requirement from the surgical classification coding dictionary can be found in the previous description and will not be repeated here.

[0172] The first number of diagnosis classification codes and the second number of surgical operation classification codes may be concatenated and arranged in order of priority to form the third text.

[0173] The priority order of the first number of diagnosis classification codes and the second number of surgical operation classification codes may, for example, be that the priority order of the diagnosis classification codes is higher than the priority order of the surgical operation classification codes. The determination of the priority order of different diagnosis classification codes and the priority order of different surgical operation classification codes can be detailed in the above description.

[0174] In addition, feature data of at least one feature type extracted from medical data may include not only feature data corresponding to the diagnosis category features and surgical operation category features described above, but may also include feature data of feature types such as age, gender, number of hospitalization days, consumption expenses, department, etc. The present application does not impose specific restrictions on this. If the feature data of these feature types are formatted data, they can be directly converted into word vectors as feature data. If they are unformatted data, they can be processed first using NLP methods, such as word segmentation and keyword extraction, and then the keywords can be converted into word vectors as feature data.

[0175] In some embodiments, based on feature data of at least one feature type, using a classification prediction model to preset a target diagnosis group category to which a target user belongs may include:

[0176] Fusing feature data of at least one feature type to obtain a fused feature;

[0177] The fused features are input into the classification prediction model to obtain the target diagnosis group category to which the target user belongs.

[0178] Optionally, feature data of at least one feature type may be weighted and fused to obtain a fused feature. The weight coefficient corresponding to each feature type may be preset.

[0179] Accordingly, when training the classification prediction model, using the medical sample data and the diagnosis grouping categories, the training of the classification prediction model may include:

[0180] Extracting feature sample data of at least one feature type from the medical sample data;

[0181] The classification prediction model is trained using feature sample data of at least one feature type and diagnostic grouping categories.

[0182] In some embodiments, extracting feature sample data of at least one feature type from the medical sample data may include:

[0183] Extracting a third number of diagnosis classification codes from the diagnosis classification code dictionary whose similarity to the medical sample data meets a first similarity requirement, and extracting a fourth number of surgical operation classification codes from the surgical operation classification code dictionary whose similarity to the medical sample data meets a second similarity requirement;

[0184] Based on the third number of diagnostic classification codes, the diagnostic category sample characteristics are obtained;

[0185] The sample characteristics of the surgical operation category are obtained based on the fourth number of diagnostic classification codes.

[0186] It should be noted that the third number is not necessarily the same as the first number described above, and the number of diagnosis classification codes extracted from different medical data that meet the first similarity requirement with the medical data is not necessarily the same. The fourth number is not necessarily the same as the second number described above, and the number of surgical operation classification codes extracted from different medical data that meet the second similarity requirement with the medical data is not necessarily the same.

[0187] As an optional implementation, obtaining diagnostic category sample features based on the first number of diagnostic classification codes may include:

[0188] Concatenate the first number of diagnostic classification codes in order of priority to obtain a first sample text;

[0189] Convert the first sample text into a text vector to obtain diagnostic category sample features;

[0190] Then, obtaining surgical operation category sample features based on the second number of diagnosis classification codes may include:

[0191] Concatenate the second number of surgical operation classification codes in order of priority to obtain a second sample text;

[0192] The second sample text is converted into a text vector to obtain the surgical operation category features.

[0193] As another optional implementation, obtaining diagnostic category sample features based on the third number of diagnostic classification codes may include:

[0194] Converting the third number of diagnostic classification codes into first feature sample vectors respectively, and fusing the third number of first feature sample vectors to obtain diagnostic category sample features;

[0195] Then, the sample features of the surgical operation category obtained based on the fourth number of diagnostic classification codes may include:

[0196] The fourth number of surgical operation classification codes are converted into second feature sample vectors respectively, and the fourth number of second feature sample vectors are fused to obtain surgical operation category sample features.

[0197] In some embodiments, fusing the third number of first feature sample vectors to obtain diagnostic category sample features may include:

[0198] Determining weight coefficients corresponding to the third number of first feature sample vectors respectively according to the priority order of the third number of diagnosis classification codes;

[0199] According to the weight coefficients respectively corresponding to the third number of first feature sample vectors, the third number of first feature sample vectors are weightedly fused to obtain diagnostic category sample features;

[0200] Then, converting the fourth number of surgical operation classification codes into second feature sample vectors respectively, and fusing the fourth number of second feature sample vectors to obtain surgical operation category sample features may include:

[0201] Determining weight coefficients corresponding to the fourth number of second feature sample vectors respectively according to the priority order of the fourth number of surgical operation classification codes;

[0202] The fourth number of second feature sample vectors are weightedly fused according to the weight coefficients respectively corresponding to the fourth number of second feature sample vectors to obtain surgical operation category sample features.

[0203] In some embodiments, extracting feature sample data of at least one feature type from the medical sample data may include:

[0204] Extracting a third number of diagnosis classification codes from the diagnosis classification code dictionary whose similarity to the medical sample data meets a first similarity requirement, and extracting a fourth number of surgical operation classification codes from the surgical operation classification code dictionary whose similarity to the medical sample data meets a second similarity requirement;

[0205] Concatenate a third number of diagnosis classification codes and a fourth number of surgical operation classification codes to form a third sample text;

[0206] The third sample text is converted into a text vector to obtain feature sample data.

[0207] In some embodiments, extracting a third number of diagnosis classification codes whose similarity to the medical sample data meets a first similarity requirement from the diagnosis classification code dictionary, and extracting a fourth number of surgical operation classification codes whose similarity to the medical sample data meets a second similarity requirement from the surgical operation classification code dictionary may include:

[0208] Acquire at least one first sample medical record text related to the diagnosis classification in the medical sample data;

[0209] Acquire at least one second sample medical record text related to the surgical operation classification in the medical sample data;

[0210] Extracting at least one first sample keyword related to the diagnosis classification from the first sample medical record text;

[0211] Extract at least one second sample keyword related to the surgical operation classification in the second sample medical record text

[0212] Performing similarity matching between the first sample keyword and the diagnosis classification code dictionary to obtain a diagnosis classification code that meets a first similarity requirement with the first sample keyword;

[0213] The second sample keyword is matched with the surgical operation classification code dictionary for similarity to obtain a surgical operation classification code that meets the first similarity requirement with the second sample keyword.

[0214] In some embodiments, using feature sample data of at least one feature type and a diagnosis grouping category, training a classification prediction model may include:

[0215] Fusing feature sample data of at least one feature type to obtain fused sample features;

[0216] The classification prediction model is trained by using the fusion sample features and the diagnostic group category to which the sample users belong.

[0217] In order to further improve the accuracy of the model, in some embodiments, fusing feature sample data of at least one feature type to obtain fused sample features may include:

[0218] Perform feature screening on at least one feature type to obtain feature sample data of at least one valid feature type;

[0219] The feature sample data of at least one valid feature type are fused to obtain a fused sample feature.

[0220] Among them, the feature screening method can be, for example, screening based on statistical methods, such as correlation coefficient method, mutual information method, chi-square test, etc.

[0221] Of course, a model-based feature screening method can also be used. For example, at least one sample feature type and a valid sample feature type marked in at least one sample feature type can be used to train a feature screening model, so that the feature screening model can determine the valid feature type from at least one feature type.

[0222] In addition, those skilled in the art can understand that when training the classification prediction model or using the classification pre-model to predict fish diagnosis groups, the fusion features or fusion sample features of the classification prediction model are input. If the vector dimensions are large, conventional processing operations such as dimensionality reduction can also be performed. In addition, when extracting feature data of at least one feature type from medical data, conventional data mining operations such as normalization, missing value filling, and one-hot encoding may be required for data of certain feature types. This application does not specifically limit this.

[0223] Among them, the classification prediction model in the embodiment of the present application can be implemented by a machine learning model, and the machine learning model includes but is not limited to logistic regression, SVM (Support Vector Machine), random forest, GBDT (Gradient Boosting Decision Tree), LightGBM (Light Gradient Boosting Machine, distributed gradient boosting framework), deep neural network, or an integrated model integrating multiple machine learning models, etc., which is not specifically limited in the present application.

[0224] As can be seen from the foregoing description, the target user involved in the embodiments of the present application may refer to a target patient to be classified, such as an inpatient or an outpatient, etc. Taking the target user as a target patient as an example, as another embodiment, the present application further provides a medical data processing method, including:

[0225] Obtain medical data of target patients;

[0226] Based on the medical data, using a classification prediction model to predict the target diagnostic group category to which the target patient belongs;

[0227] The classification prediction model is pre-trained using the medical sample data corresponding to the sample patients and the diagnostic grouping categories to which they belong.

[0228] This embodiment and Figure 1 The difference between the illustrated embodiments is that the target user is specifically a target patient, and the sample user is specifically a sample patient, which is used to predict the target diagnostic group category to which the target patient belongs. Other identical or similar steps can be found in detail in the previous embodiments and will not be repeated here.

[0229] As another embodiment, the present application also provides a medical data processing method, including:

[0230] Determine the medical sample data of the sample patients and the diagnostic group category to which they belong;

[0231] The classification prediction model is trained using the medical sample data and the diagnostic grouping categories.

[0232] This embodiment is described from the perspective of model training. Figure 2 The difference between the illustrated embodiments is that the sample users specifically refer to sample patients, and other identical or similar steps can be found in the foregoing embodiments and will not be described in detail here.

[0233] As another embodiment, the present application also provides a medical data processing method, including:

[0234] Obtain medical data of target patients;

[0235] Based on the medical data, using a classification prediction model to predict the target diagnostic group category to which the target patient belongs;

[0236] Execute corresponding task processing operations according to the target diagnosis group category;

[0237] The classification prediction model is pre-trained using the medical sample data corresponding to the sample patients and the diagnostic grouping categories to which they belong.

[0238] This embodiment and Figure 3 The difference between the illustrated embodiments is that the target user is specifically a target patient, and the sample user is specifically a sample patient. Other identical or similar steps can be found in the description of the previous embodiments and will not be repeated here.

[0239] In a practical application, the technical solution of the embodiment of the present application can be specifically used to group inpatients with DRGs. The DRGs grouping category of inpatients can be predicted based on the data on the first page of the medical record filled out by the inpatients when they are discharged from the hospital, thereby eliminating the need for manual grouping by professionals, improving the efficiency of diagnostic grouping and ensuring the accuracy of diagnostic grouping, and then based on the DRGs grouping category, the medical institution where the inpatients are located can be compensated for expenses, such as medical insurance reimbursement, or service evaluation, such as performance appraisal, can be performed.

[0240] In addition, the DRGs grouping category to which the inpatient belongs can also be dynamically predicted based on the medical record data generated in real time during the inpatient treatment process. Therefore, treatment recommendation information can be generated based on the DRGs grouping category to which the inpatient belongs to decide whether to adjust the inpatient's medical plan.

[0241] Take the dynamic prediction of DRGs grouping categories of hospitalized patients as an example. Figure 5 As shown in the training and application diagram of the classification prediction model shown in , the classification prediction model can be first trained in combination with the medical sample data of the sample users and the DRGs grouping categories to which they belong. The sample users may refer to historical inpatients of the medical institution, and the medical sample data of the historical inpatients can be obtained from the historical data records in the medical data system 501 of the medical institution. Thus, the classification prediction model can be trained by utilizing a large amount of medical sample data of historical inpatients and the DRGs grouping categories to which they belong.

[0242] The classification prediction model obtained through training can be used to dynamically predict the current inpatients in the medical institution. The medical record data of the current inpatients can be obtained from the medical data system 501, and the medical record data can be input into the classification prediction model to obtain the target DRGs grouping category to which the current inpatients belong. Based on the DRGs grouping category, corresponding task processing operations can be performed, such as generating diagnosis and treatment recommendation information, and the diagnosis and treatment recommendation information can be provided to professionals, etc.

[0243] Among them, the training of the classification prediction model can be implemented by a dedicated computing device 502, such as the computing device can refer to a remote web server, server, computer networking device, chipset, desktop computer, notebook computer, workstation, or any other processing device or equipment.

[0244] The generated classification prediction model can be integrated into the current medical data system 501. When the medical data system detects that the medical record data of the current inpatient is updated, the classification prediction model can be used to predict the DRGs grouping category to which the current inpatient belongs. Of course, the DRGs grouping category can also be predicted for multiple inpatients in batches.

[0245] Of course, in an optional manner, the computing device 502 can be integrated into the medical data system 501, so that the medical data system can use its own historical data to complete model training, and can perform diagnostic grouping predictions on current inpatients, etc. The present application does not specifically limit the execution device for implementing classification prediction model training and the execution device for performing diagnostic grouping based on the classification prediction model.

[0246] By adopting the technical solution of the present application to perform DRGs grouping, it is possible to implement DRGs grouping without knowing the DRGs standard, and it is possible to save the grouping operation by professionals. Each medical institution can perform predictions after deploying the classification prediction model, thereby improving the efficiency of DRGs grouping. Moreover, each medical institution can implement DRGs grouping predictions during the patient's hospitalization process, so that the patient's medical plan can be adjusted according to actual needs, etc., which is convenient for carrying out DRGs related work. Moreover, the classification prediction model is applicable to different DRGs standards. It only needs to input the corresponding training data of the corresponding DRGs standard (including medical sample data and the DRGs grouping category under the DRGs standard) during model training, and it has strong versatility.

[0247] Figure 6 A schematic diagram of a structure of an embodiment of a data processing device provided in an embodiment of the present application, the data processing device may be configured in a computing device, and the device may include:

[0248] The first acquisition module 601 is used to acquire the medical data of the target user;

[0249] Category prediction module 602, used to predict the target diagnosis group category to which the target user belongs based on the medical data using a classification prediction model;

[0250] Among them, the classification prediction model is pre-trained using the medical sample data corresponding to the sample users and the diagnostic group categories to which they belong.

[0251] In some embodiments, the first acquisition module may be specifically used to acquire the medical record data generated by the target user during the diagnosis and treatment process.

[0252] Optionally, the first acquisition module may specifically acquire the medical record data of the target user from the medical data system when the medical data system updates the medical record of the target user during the diagnosis and treatment process of the target user.

[0253] In some embodiments, the first acquisition module may specifically acquire the medical record homepage data from the medical record homepage corresponding to the target user.

[0254] In some embodiments, the apparatus may further include:

[0255] The task processing module is used to perform corresponding task processing operations according to the target diagnosis group category.

[0256] As an optional method, the task processing module can be specifically used to conduct service evaluation or fee compensation for the medical institution where the target user is located according to the target diagnosis group category.

[0257] As another optional manner, the task processing module may be specifically used to generate diagnosis and treatment recommendation information for a target user according to the target diagnosis group category.

[0258] Optionally, the task processing module may specifically generate diagnosis and treatment recommendation information for the target user in combination with the target diagnosis group category and relevant medical attribute information of the target user.

[0259] In some embodiments, the medical attribute information includes the target user's actual diagnosis, target department, and / or current consumption information;

[0260] This task processing module can be specifically used to generate diagnosis and treatment recommendation information for the target user if the diagnosis result corresponding to the target diagnosis grouping category is different from the actual diagnosis result, the target diagnosis grouping category is different from the diagnosis grouping category corresponding to the target department where the target user is located, and / or the consumption cost corresponding to the target diagnosis grouping category is inconsistent with the current consumption cost of the target user.

[0261] In some embodiments, the category prediction module can be specifically used to extract feature data of at least one feature type from medical data; based on the feature data of at least one feature type, use a classification prediction model to identify the target diagnosis group category to which the target user belongs.

[0262] In some embodiments, the category prediction module extracts feature data of at least one feature type from medical data, which may include: extracting a first number of diagnosis classification codes from a diagnosis classification code dictionary whose similarity to the medical data meets a first similarity requirement, and extracting a second number of surgical operation classification codes from a surgical classification code dictionary whose similarity to the medical data meets a second similarity requirement; obtaining diagnosis category features based on the first number of diagnosis classification codes; and obtaining surgical operation category features based on the second number of diagnosis classification codes.

[0263] In some embodiments, the category prediction module obtains the diagnosis category feature based on the first number of diagnosis category codes, including: splicing the first number of diagnosis category codes in order of priority to obtain a first text; converting the first text into a text vector to obtain the diagnosis category feature;

[0264] The category prediction module obtains surgical operation category features based on the second number of diagnosis classification codes, including: concatenating the second number of surgical operation classification codes in order of priority to obtain a second text; converting the second text into a text vector to obtain surgical operation category features.

[0265] In some embodiments, the category prediction module obtains the diagnosis category feature based on the first number of diagnosis category codes, including: converting the first number of diagnosis category codes into first feature vectors respectively, and fusing the first number of first feature vectors to obtain the diagnosis category feature;

[0266] The prediction module obtains surgical operation category features based on the second number of surgical operation classification codes, including: converting the second number of surgical operation classification codes into second feature vectors respectively, and fusing the second number of second feature vectors to obtain surgical operation category features.

[0267] In some embodiments, the category prediction module fuses the first number of first feature vectors to obtain the diagnosis category feature, including: determining the weight coefficients corresponding to the first number of first feature vectors according to the priority order of the first number of diagnosis classification codes; weightedly fusing the first number of first feature vectors according to the weight coefficients corresponding to the first number of first feature vectors to obtain the diagnosis category feature;

[0268] The category prediction module fuses the second number of second feature vectors to obtain the surgical operation category features, including: determining the weight coefficients corresponding to the second number of second feature vectors according to the priority order of the second number of surgical operation classification codes; and weightedly fusing the second number of second feature vectors according to the weight coefficients corresponding to the second number of second feature vectors to obtain the surgical operation category features.

[0269] In some embodiments, the category prediction module extracts feature data of at least one feature type from the medical data, including: extracting a first number of diagnosis classification codes whose similarity to the medical data meets a first similarity requirement from a diagnosis classification code dictionary, and extracting a second number of surgical operation classification codes whose similarity to the medical data meets a second similarity requirement from a surgical classification code dictionary; concatenating the first number of diagnosis classification codes and the second number of surgical operation classification codes to form a third text; converting the third text into a text vector to obtain feature data.

[0270] In certain embodiments, the category prediction module extracts a first number of diagnosis classification codes whose similarity with the medical data meets a first similarity requirement from the diagnosis classification code dictionary, and extracts a second number of surgical operation classification codes whose similarity with the medical data meets a second similarity requirement from the surgical classification code dictionary, including: obtaining at least one first medical record text related to the diagnosis classification in the medical data; obtaining at least one second medical record text related to the surgical operation classification in the medical data; extracting at least one first keyword related to the diagnosis classification in the first medical record text; extracting at least one second keyword related to the surgical operation classification in the second medical record text; performing similarity matching between the first keyword and the diagnosis classification code dictionary to obtain a diagnosis classification code that meets the first similarity requirement with the first keyword; performing similarity matching between the second keyword and the surgical operation classification code dictionary to obtain a surgical operation classification code that meets the second similarity requirement with the second keyword.

[0271] In some embodiments, the category prediction module presets the target diagnostic group category to which the target user belongs based on feature data of at least one feature type using a classification prediction model, including: fusing feature data of at least one feature type to obtain a fused feature; and inputting the fused feature into the classification prediction model to obtain the target diagnostic group category to which the target user belongs.

[0272] In a practical application, the diagnostic grouping category in the embodiment of the present application may refer to the diagnosis-related classification DRGs grouping category.

[0273] Figure 6 The data processing device can execute Figure 1The implementation principle and technical effect of the data processing method described in the illustrated embodiment will not be described in detail. The specific manner in which each module and unit performs operations in the data processing device in the above embodiment has been described in detail in the embodiment of the method, and will not be described in detail here.

[0274] In one possible design, Figure 6 The data processing apparatus of the illustrated embodiment may be implemented as a computing device, such as Figure 7 As shown in , the computing device may include a storage component 701 and a processing component 702;

[0275] The storage component 701 stores one or more computer instructions, wherein the one or more computer instructions are called and executed by the processing component 702 to implement the following Figure 1 or Figure 3 The data processing method shown.

[0276] Of course, a computing device may also include other components, such as input / output interfaces, communication components, etc.

[0277] The input / output interface provides an interface between the processing component and the peripheral interface module, which may be an output device, an input device, etc.

[0278] The communication component is configured to facilitate, among other things, wired or wireless communications between the computing device and other devices.

[0279] The computing device may be a physical device or an elastic computing host provided by a cloud computing platform, etc. In this case, the computing device may refer to a cloud server, and the above-mentioned processing components, storage components, etc. may be basic server resources rented or purchased from the cloud computing platform.

[0280] The present application also provides a computer-readable storage medium storing a computer program. When the computer program is executed by a computer, the above-mentioned Figure 1 or Figure 3 The data processing method of the illustrated embodiment.

[0281] Figure 8 A schematic diagram of a structure of another embodiment of a data processing device provided in an embodiment of the present application, the data processing device may be configured in a computing device, and the device may include:

[0282] The second acquisition module 801 is used to determine the medical sample data of the sample user and the diagnostic group category to which it belongs;

[0283] The model training module 802 is used to train a classification prediction model using medical sample data and diagnostic grouping categories.

[0284] In some embodiments, the model training module is specifically used to extract feature sample data of at least one feature type from the medical sample data; and train the classification prediction model using the feature sample data of at least one feature type and the diagnostic grouping category.

[0285] In certain embodiments, the model training module extracts feature sample data of at least one feature type from the medical sample data, including: extracting a third number of diagnosis classification codes from a diagnosis classification code dictionary whose similarity with the medical sample data meets a first similarity requirement, and extracting a fourth number of surgical operation classification codes from a surgical classification code dictionary whose similarity with the medical sample data meets a second similarity requirement; obtaining diagnosis category sample features based on the third number of diagnosis classification codes; and obtaining surgical operation category sample features based on the fourth number of diagnosis classification codes.

[0286] In some embodiments, the model training module obtains the diagnostic category sample features based on the first number of diagnostic category codes, including: splicing the first number of diagnostic category codes in order of priority to obtain a first sample text; converting the first sample text into a text vector to obtain the diagnostic category sample features;

[0287] The model training module obtains surgical operation category sample features based on the second number of diagnosis classification codes, including: concatenating the second number of surgical operation classification codes in order of priority to obtain a second sample text; converting the second sample text into a text vector to obtain surgical operation category features.

[0288] In some embodiments, the model training module obtains the diagnostic category sample features based on the third number of diagnostic category codes, including: converting the third number of diagnostic category codes into first feature sample vectors respectively, and fusing the third number of first feature sample vectors to obtain the diagnostic category sample features;

[0289] The model training module obtains surgical operation category sample features based on the fourth number of diagnosis classification codes, including: converting the fourth number of surgical operation classification codes into second feature sample vectors respectively, and fusing the fourth number of second feature sample vectors to obtain surgical operation category sample features.

[0290] In some embodiments, the model training module fuses the third number of first feature sample vectors to obtain the diagnostic category sample features, including: determining the weight coefficients corresponding to the third number of first feature sample vectors according to the priority order of the third number of diagnostic classification codes; weightedly fusing the third number of first feature sample vectors according to the weight coefficients corresponding to the third number of first feature sample vectors to obtain the diagnostic category sample features;

[0291] The model training module converts the fourth number of surgical operation classification codes into second feature sample vectors respectively, and fuses the fourth number of second feature sample vectors to obtain surgical operation category sample features, including: determining the weight coefficients corresponding to the fourth number of second feature sample vectors respectively according to the priority order of the fourth number of surgical operation classification codes; and weightedly fusing the fourth number of second feature sample vectors according to the weight coefficients corresponding to the fourth number of second feature sample vectors to obtain surgical operation category sample features.

[0292] In some embodiments, the model training module extracts feature sample data of at least one feature type from the medical sample data, including: extracting a third number of diagnosis classification codes from a diagnosis classification code dictionary whose similarity with the medical sample data meets a first similarity requirement, and extracting a fourth number of surgical operation classification codes from a surgical classification code dictionary whose similarity with the medical sample data meets a second similarity requirement; concatenating the third number of diagnosis classification codes and the fourth number of surgical operation classification codes to form a third sample text; converting the third sample text into a text vector to obtain feature sample data.

[0293] In certain embodiments, the model training module extracts a third number of diagnosis classification codes whose similarity with the medical sample data meets a first similarity requirement from the diagnosis classification code dictionary, and extracts a fourth number of surgical operation classification codes whose similarity with the medical sample data meets a second similarity requirement from the surgical operation classification code dictionary, including: obtaining at least one first sample medical record text related to the diagnosis classification in the medical sample data; obtaining at least one second sample medical record text related to the surgical operation classification in the medical sample data; extracting at least one first sample keyword related to the diagnosis classification in the first sample medical record text; extracting at least one second sample keyword related to the surgical operation classification in the second sample medical record text; performing similarity matching between the first sample keyword and the diagnosis classification code dictionary to obtain a diagnosis classification code that meets the first similarity requirement with the first sample keyword; performing similarity matching between the second sample keyword and the surgical operation classification code dictionary to obtain a surgical operation classification code that meets the first similarity requirement with the second sample keyword.

[0294] In some embodiments, the model training module uses feature sample data of at least one feature type and a diagnostic grouping category to train a classification prediction model, including: fusing feature sample data of at least one feature type to obtain fused sample features; and using the fused sample features and the diagnostic grouping category to train the classification prediction model.

[0295] In some embodiments, the model training module fuses feature sample data of at least one feature type to obtain fused sample features, including: performing feature screening on at least one feature type to obtain feature sample data of at least one valid feature type; and fusing feature sample data of at least one valid feature type to obtain fused sample features.

[0296] Figure 8 The data processing device can execute Figure 2 The implementation principle and technical effect of the data processing method described in the illustrated embodiment will not be described in detail. The specific manner in which each module and unit performs operations in the data processing device in the above embodiment has been described in detail in the embodiment of the method, and will not be described in detail here.

[0297] In one possible design, Figure 8 The data processing apparatus of the illustrated embodiment may be implemented as a computing device, such as Fig. 9 As shown in , the computing device may include a storage component 901 and a processing component 902;

[0298] The storage component 901 stores one or more computer instructions, wherein the one or more computer instructions are called and executed by the processing component 902 to implement the following Figure 1 or Figure 3 The data processing method shown.

[0299] Of course, a computing device may also include other components, such as input / output interfaces, communication components, etc.

[0300] The input / output interface provides an interface between the processing component and the peripheral interface module, which may be an output device, an input device, etc.

[0301] The communication component is configured to facilitate, among other things, wired or wireless communications between the computing device and other devices.

[0302] The computing device may be a physical device or an elastic computing host provided by a cloud computing platform, etc. In this case, the computing device may refer to a cloud server, and the above-mentioned processing components, storage components, etc. may be basic server resources rented or purchased from the cloud computing platform.

[0303] Fig. 9 The computing device shown is Figure 8 The computing device shown can be the same computing device in actual applications, and can also be different computing devices to respectively implement the training of the classification prediction model and the operation of performing diagnosis grouping using the classification prediction model.

[0304] The present application also provides a computer-readable storage medium storing a computer program. When the computer program is executed by a computer, the above-mentioned Figure 2 The data processing method of the illustrated embodiment.

[0305] The processing components in the above-mentioned embodiments may include one or more processors to execute computer instructions to complete all or part of the steps in the above-mentioned method. Of course, the processing components may also be implemented by one or more application-specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field programmable gate arrays (FPGAs), controllers, microcontrollers, microprocessors or other electronic components to perform the above-mentioned methods.

[0306] The storage component is configured to store various types of data to support operations at the terminal. The storage component can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic disk or optical disk.

[0307] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0308] The device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components displayed 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. Ordinary technicians in this field can understand and implement it without paying creative labor.

[0309] Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a necessary general hardware platform, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solution is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a 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 device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0310] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit it. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A medical data processing method, characterized in that: include: Obtain medical data of target patients; extracting feature data of at least one feature type from the medical data; Fusing the feature data of the at least one feature type to obtain a fused feature; Inputting the fusion features into a classification prediction model to obtain a target diagnostic grouping category to which the target patient belongs; The classification prediction model is pre-trained using the medical sample data corresponding to the sample patients and the diagnostic grouping categories to which they belong.

2. A medical data processing method, characterized in that: include: Determine the medical sample data of the sample patients and the diagnostic group category to which they belong; Extracting feature sample data of at least one feature type from the medical sample data; Fusing the feature sample data of the at least one feature type to obtain a fused sample feature; The classification prediction model is trained using the fused sample features and the diagnostic grouping categories; wherein the classification prediction model is used to extract feature data of at least one feature type from the medical data based on the medical data of a target patient, fuse the feature data of the at least one feature type to obtain a fused feature, and obtain the target diagnostic grouping category to which the target patient belongs based on the fused feature.

3. A medical data processing method, characterized in that: include: Obtain medical data of target patients; extracting feature data of at least one feature type from the medical data; Fusing the feature data of the at least one feature type to obtain a fused feature; Inputting the fusion features into a classification prediction model to obtain a target diagnostic grouping category to which the target patient belongs; Execute corresponding task processing operations according to the target diagnosis group category; The classification prediction model is pre-trained using the medical sample data corresponding to the sample patients and the diagnostic grouping categories to which they belong.

4. A data processing method, characterized in that: include: Obtain medical data of target users; extracting feature data of at least one feature type from the medical data; Fusing the feature data of the at least one feature type to obtain a fused feature; Inputting the fusion features into a classification prediction model to obtain a target diagnosis group category to which the target user belongs; The classification prediction model is pre-trained using the medical sample data corresponding to the sample users and the diagnostic grouping categories to which they belong.

5. The method according to claim 4, characterized in that The obtaining of the target user's medical data includes: Obtain the medical record data generated by the target user during the diagnosis and treatment process.

6. The method according to claim 5, characterized in that The acquisition of the medical record data generated by the target user during the diagnosis and treatment process includes: When the medical data system updates the medical record of the target user during the diagnosis and treatment of the target user, the medical record data of the target user is obtained from the medical data system.

7. The method according to claim 4, characterized in that The obtaining of the target user's medical data includes: The medical record homepage data is obtained from the medical record homepage corresponding to the target user.

8. The method according to claim 4, characterized in that The method further comprises: According to the target diagnosis grouping category, a corresponding task processing operation is performed.

9. The method according to claim 8, characterized in that The performing of corresponding task processing operations according to the target diagnosis grouping category includes: According to the target diagnosis grouping category, a service evaluation or fee compensation is performed on the medical institution where the target user is located.

10. The method according to claim 8, characterized in that The performing of corresponding task processing operations according to the target diagnosis grouping category includes: According to the target diagnosis grouping category, diagnosis and treatment recommendation information for the target user is generated.

11. The method according to claim 10, characterized in that The generating of diagnosis and treatment suggestion information for the target user according to the target diagnosis group category includes: Combine the target diagnosis group category and the relevant medical attribute information of the target user to generate diagnosis and treatment recommendation information for the target user.

12. The method according to claim 11, characterized in that The medical attribute information includes the actual diagnosis result, target department and / or current consumption information of the target user; The generating of diagnosis and treatment suggestion information for the target user according to the target diagnosis group category and the medical attribute information related to the target user includes: If the diagnosis result corresponding to the target diagnostic grouping category is different from the actual diagnosis result, the target diagnostic grouping category is different from the diagnosis grouping category corresponding to the target department where the target user is located, and / or the consumption cost corresponding to the target diagnostic grouping category is inconsistent with the current consumption cost of the target user, generate diagnosis and treatment recommendation information for the target user.

13. The method according to claim 4, characterized in that The extracting feature data of at least one feature type from the medical data comprises: Extracting a first number of diagnosis classification codes whose similarity to the medical data meets a first similarity requirement from a diagnosis classification code dictionary, and extracting a second number of surgical operation classification codes whose similarity to the medical data meets a second similarity requirement from a surgical operation classification code dictionary; Obtaining diagnostic category characteristics based on the first number of diagnostic classification codes; A surgical operation category feature is obtained based on the second number of diagnosis classification codes.

14. The method according to claim 13, characterized in that The obtaining of the diagnosis category features based on the first number of diagnosis classification codes comprises: Concatenate the first number of diagnostic classification codes in order of priority to obtain a first text; Convert the first text into a text vector to obtain a diagnostic category feature; The obtaining of surgical operation category features based on the second number of diagnosis classification codes includes: concatenating the second number of surgical operation classification codes in order of priority to obtain a second text; The second text is converted into a text vector to obtain surgical operation category features.

15. The method according to claim 13, characterized in that The obtaining of the diagnosis category features based on the first number of diagnosis classification codes comprises: Converting the first number of diagnosis classification codes into first feature vectors respectively, and fusing the first number of first feature vectors to obtain diagnosis category features; The obtaining of surgical operation category features based on the second number of surgical operation classification codes comprises: The second number of surgical operation classification codes are respectively converted into second feature vectors, and the second number of second feature vectors are fused to obtain surgical operation category features.

16. The method according to claim 15, characterized in that The fusing the first number of first feature vectors to obtain the diagnostic category feature includes: Determining weight coefficients corresponding to the first number of first feature vectors respectively according to the priority order of the first number of diagnosis classification codes; According to the weight coefficients respectively corresponding to the first number of first feature vectors, weighted fusion of the first number of first feature vectors is performed to obtain a diagnostic category feature; The step of fusing the second number of second feature vectors to obtain the surgical operation category feature comprises: Determining weight coefficients corresponding to the second number of second feature vectors respectively according to the priority order of the second number of surgical operation classification codes; The second number of second feature vectors are weightedly fused according to the weight coefficients respectively corresponding to the second number of second feature vectors to obtain the surgical operation category feature.

17. The method according to claim 4, characterized in that Extracting feature data of at least one feature type from the medical data includes: Extracting a first number of diagnosis classification codes whose similarity to the medical data meets a first similarity requirement from a diagnosis classification code dictionary, and extracting a second number of surgical operation classification codes whose similarity to the medical data meets a second similarity requirement from a surgical operation classification code dictionary; splicing the first number of diagnosis classification codes and the second number of surgical operation classification codes to form a third text; The third text is converted into a text vector to obtain feature data.

18. The method according to claim 13 or 17, characterized in that The extracting of a first number of diagnosis classification codes whose similarity to the medical data meets a first similarity requirement from the diagnosis classification code dictionary, and extracting a second number of surgical operation classification codes whose similarity to the medical data meets a second similarity requirement from the surgical operation classification code dictionary includes: Acquire at least one first medical record text related to the diagnosis classification in the medical data; Acquire at least one second medical record text related to the surgical operation classification in the medical data; Extracting at least one first keyword related to the diagnosis classification from the first medical record text; Extracting at least one second keyword related to the surgical operation classification from the second medical record text; Performing similarity matching between the first keyword and a diagnosis classification code dictionary to obtain a diagnosis classification code that meets a first similarity requirement with the first keyword; The second keyword is matched with a surgical operation classification code dictionary for similarity to obtain a surgical operation classification code that meets a second similarity requirement with the second keyword.

19. The method according to claim 4, characterized in that The diagnostic grouping categories are specifically diagnosis-related classification DRGs grouping categories.

20. A data processing method, characterized in that: include: Determine the medical sample data of the sample user and the diagnostic group category to which it belongs; Extracting feature sample data of at least one feature type from the medical sample data; Fusing the feature sample data of the at least one feature type to obtain a fused sample feature; The classification prediction model is trained using the fused sample features and the diagnostic grouping categories; wherein the classification prediction model is used to extract feature data of at least one feature type from the medical data based on the target user's medical data, fuse the feature data of the at least one feature type to obtain a fused feature, and obtain the target diagnostic grouping category to which the target user belongs based on the fused feature.

21. The method according to claim 20, characterized in that The extracting feature sample data of at least one feature type from the medical sample data comprises: Extracting a third number of diagnosis classification codes whose similarity to the medical sample data meets a first similarity requirement from the diagnosis classification code dictionary, and extracting a fourth number of surgical operation classification codes whose similarity to the medical sample data meets a second similarity requirement from the surgical operation classification code dictionary; Obtaining diagnostic category sample characteristics based on the third number of diagnostic classification codes; The surgical operation category sample features are obtained based on the fourth number of diagnosis classification codes.

22. The method according to claim 21, characterized in that The obtaining of diagnostic category sample features based on the third number of diagnostic classification codes comprises: Concatenate the third number of diagnostic classification codes in order of priority to obtain a first sample text; Convert the first sample text into a text vector to obtain diagnostic category sample features; The obtaining of surgical operation category sample features based on the fourth number of diagnosis classification codes includes: Concatenate the fourth number of surgical operation classification codes in order of priority to obtain a second sample text; The second sample text is converted into a text vector to obtain surgical operation category features.

23. The method according to claim 21, characterized in that The obtaining of diagnostic category sample features based on the third number of diagnostic classification codes comprises: Converting the third number of diagnosis classification codes into first feature sample vectors respectively, and fusing the third number of first feature sample vectors to obtain diagnosis category sample features; The obtaining of surgical operation category sample features based on the fourth number of diagnosis classification codes includes: The fourth number of surgical operation classification codes are converted into second feature sample vectors respectively, and the fourth number of second feature sample vectors are fused to obtain surgical operation category sample features.

24. The method according to claim 23, characterized in that The fusing of the third number of first feature sample vectors to obtain diagnostic category sample features includes: Determining weight coefficients corresponding to the third number of first feature sample vectors respectively according to the priority order of the third number of diagnosis classification codes; According to the weight coefficients respectively corresponding to the third number of first feature sample vectors, the third number of first feature sample vectors are weightedly fused to obtain diagnostic category sample features; The converting the fourth number of surgical operation classification codes into second feature sample vectors respectively, and fusing the fourth number of second feature sample vectors to obtain surgical operation category sample features includes: Determining weight coefficients corresponding to the fourth number of second feature sample vectors respectively according to the priority order of the fourth number of surgical operation classification codes; The fourth number of second feature sample vectors are weightedly fused according to the weight coefficients respectively corresponding to the fourth number of second feature sample vectors to obtain surgical operation category sample features.

25. The method according to claim 20, characterized in that The extracting feature sample data of at least one feature type from the medical sample data comprises: Extracting a third number of diagnosis classification codes whose similarity to the medical sample data meets a first similarity requirement from the diagnosis classification code dictionary, and extracting a fourth number of surgical operation classification codes whose similarity to the medical sample data meets a second similarity requirement from the surgical operation classification code dictionary; Concatenate the third number of diagnosis classification codes and the fourth number of surgical operation classification codes to form a third sample text; The third sample text is converted into a text vector to obtain feature sample data.

26. The method according to claim 21 or 25, characterized in that The extracting of a third number of diagnosis classification codes whose similarity to the medical sample data meets a first similarity requirement from the diagnosis classification code dictionary, and extracting a fourth number of surgical operation classification codes whose similarity to the medical sample data meets a second similarity requirement from the surgical operation classification code dictionary includes: Acquire at least one first sample medical record text related to the diagnosis classification in the medical sample data; Acquire at least one second sample medical record text related to the surgical operation classification in the medical sample data; extracting at least one first sample keyword related to the diagnosis classification from the first sample medical record text; extracting at least one second sample keyword related to the surgical operation classification from the second sample medical record text; Performing similarity matching between the first sample keyword and a diagnosis classification code dictionary to obtain a diagnosis classification code that meets a first similarity requirement with the first sample keyword; The second sample keyword is matched with a surgical operation classification code dictionary for similarity to obtain a surgical operation classification code that meets a first similarity requirement with the second sample keyword.

27. The method according to claim 20, characterized in that The fusing the feature sample data of the at least one feature type to obtain the fused sample feature comprises: Performing feature screening on the at least one feature type to obtain feature sample data of at least one valid feature type; The feature sample data of the at least one valid feature type are fused to obtain a fused sample feature.

28. A data processing method, characterized in that: include: Obtain medical data of target users; extracting feature data of at least one feature type from the medical data; Fusing the feature data of the at least one feature type to obtain a fused feature; Inputting the fusion features into a classification prediction model to obtain a target diagnosis group category to which the target user belongs; Execute corresponding task processing operations according to the target diagnosis group category; The classification prediction model is pre-trained using the medical sample data corresponding to the sample users and the diagnostic grouping categories to which they belong.

29. A data processing device, characterized in that: include: A first acquisition module is used to acquire medical data of a target user; A category prediction module, used to extract feature data of at least one feature type from the medical data; fuse the feature data of at least one feature type to obtain a fused feature; input the fused feature into a classification prediction model to obtain a target diagnosis group category to which the target user belongs; The classification prediction model is pre-trained using the medical sample data corresponding to the sample users and the diagnostic grouping categories to which they belong.

30. A data processing device, characterized in that: include: The second acquisition module is used to determine the medical sample data of the sample user and the diagnostic group category to which it belongs; A model training module is used to extract feature sample data of at least one feature type from the medical sample data; fuse the feature sample data of at least one feature type to obtain a fused sample feature; and train a classification prediction model using the fused sample feature and the diagnostic group category; wherein the classification prediction model is used to extract feature data of at least one feature type from the medical data based on the medical data of a target user, fuse the feature data of at least one feature type to obtain a fused feature, and obtain the target diagnostic group category to which the target user belongs based on the fused feature.

31. A computing device, characterized in that including a processing component and a storage component; The storage component stores one or more computer instructions; the one or more computer instructions are used to be called and executed by the processing component to implement the data processing method as described in any one of claims 4 to 19.

32. A computing device, characterized in that including a processing component and a storage component; The storage component stores one or more computer instructions; the one or more computer instructions are used to be called and executed by the processing component to implement the data processing method as described in any one of claims 20 to 28.

33. A computer storage medium, characterized in that A computer program is stored, and when the computer program is executed by a computer, the data processing method according to any one of claims 4 to 19 is implemented.

34. A computer storage medium, characterized in that A computer program is stored, and when the computer program is executed by a computer, the data processing method according to any one of claims 20 to 28 is implemented.

Images