Medical information query method and device, electronic equipment and program product

Abstract

The invention provides a medical information query method and device, electronic equipment and a program product, and the method comprises the steps: obtaining the description information of a target object for querying a drug or a disease, inputting the description information into a medical information query model, and obtaining a query result; the medical information query model is obtained by training a preset information query model through first sample data, and the first sample data is obtained by integrating multi-source medical data; the multi-source medical data are integrated to obtain the first sample data, the preset information query model is trained through the first sample data to obtain the medical information query model, and the medical information query model fuses the multi-source medical data, so that after the description information is input, the multi-source medical data can be obtained; the correlation between the output query result and the description information is high and more comprehensive, and the reference of the query result is improved.

Description

Technical Field

[0001] The present invention relates to the field of medical data processing technology, and in particular to a medical information query method, device, electronic equipment and program product. Background Art

[0002] With the improvement of living standards, people pay more and more attention to medical information. For example, people will pay attention to the efficacy of a new drug, whether it has side effects, the usage of a medical device, the cost of use, etc. at any time and anywhere. In addition, people can also enter the name of a disease into the corresponding application software on the electronic terminal to query the symptoms of the disease, etc. When feeling unwell, they can enter the physical symptoms into the corresponding application software on the electronic terminal to query guidance and suggestions.

[0003] However, the guidance suggestions given by the current application software have problems such as incomplete information and low relevance, which leads to poor reference value of the guidance suggestions. Therefore, it is necessary to improve the existing medical information query method. Summary of the invention

[0004] In view of the above-mentioned shortcomings of the prior art, the present application provides a medical information query method, device, electronic device and program product to solve the above-mentioned technical problems.

[0005] According to one aspect of an embodiment of the present application, a medical information query method is provided, including: obtaining descriptive information of a target object querying drugs or symptoms; inputting the descriptive information into a medical information query model to obtain a query result; the medical information query model is obtained by training a preset information query model with first sample data, and the first sample data is obtained by integrating multi-source medical data, and the multi-source medical data includes electronic medical record data, drug sales data, drug testing data, and medication efficacy feedback data.

[0006] In one embodiment of the present application, if the first sample data includes preset subject information and medical information associated with the preset subject information, the process of training the preset information query model through the first sample data to obtain the medical information query model includes: dividing the preset subject information into sample subject information and test subject information; training the preset information query model through the sample subject information and the medical information associated with the sample subject information to obtain a trained information query model; inputting the test subject information into the trained information query model to obtain test medical information; if the error value between the test medical information and the medical information associated with the test subject information is greater than a preset error threshold, adjusting the parameters in the trained information query model until the error value between the test medical information output by the adjusted information query model and the medical information associated with the test subject information is less than or equal to the preset error threshold, thereby obtaining the medical information query model.

[0007] In one embodiment of the present application, the process of integrating multi-source medical data to obtain the first sample data includes: preprocessing the electronic medical record data, the drug sales data, the drug testing data and the medication efficacy feedback data to obtain drug usage data; the preprocessing method includes missing data filling, data deduplication, data denoising, data format conversion and data normalization; extracting feature information of the drug usage data to obtain multiple feature information; associating multiple feature information according to preset subject information to obtain multiple associated feature information; the preset subject information includes identity card number, and / or drug number; and using the multiple associated feature information as the first sample data.

[0008] In one embodiment of the present application, the process of integrating multi-source medical data to obtain the first sample data includes: preprocessing the electronic medical record data, the drug sales data, the drug testing data and the medication efficacy feedback data to obtain drug usage data; the preprocessing method includes missing data filling, data deduplication, data denoising, data format conversion and data normalization; extracting feature information of the drug usage data to obtain multiple feature information; associating multiple feature information according to preset subject information to obtain multiple associated feature information; the preset subject information includes identity card number, and / or drug number; according to preset filtering dimensions, filtering multiple associated feature information to obtain a filtered information set; using the data in the filtered information set as the first sample data.

[0009] In one embodiment of the present application, according to the preset classification dimension and the preset filtering dimension, multiple pieces of the related characteristic information are filtered to obtain the filtered information set, including: if the preset subject information also includes age information, gender information, disease type and residential area, then at least one of the age information, the gender information, the disease type and the residential area is used as the preset filtering dimension to filter the multiple pieces of the related characteristic information to obtain the filtered information set; if the preset subject information also includes sales location and usage method, then at least one of the sales location and the usage method is used as the preset filtering dimension to filter the multiple pieces of the related characteristic information to obtain the filtered information set.

[0010] In one embodiment of the present application, after obtaining the screening information set, the method also includes: if the number of the screening information sets is less than a preset number threshold, fitting the data in the screening information set to obtain a screening feature fitting curve; if the number of the screening information sets is greater than or equal to the preset number threshold, fitting the data in each screening information set to obtain multiple screening feature fitting curves, and comparing the multiple screening feature fitting curves to obtain data differences; if the data difference is greater than or equal to a preset difference value, determining that the preset screening dimension has an influence relationship on the medical information associated with the preset subject information; if the data difference is less than the preset difference value, determining that the preset screening dimension has no influence relationship on the medical information associated with the preset subject information.

[0011] In one embodiment of the present application, before extracting the characteristic information from the drug usage data, the method also includes: taking the drug usage data within a preset time period as second sample data, and performing feature annotation on the second sample data to obtain annotated data; training a preset feature extraction module through the annotated data to obtain a first feature extraction module; and training a preset associated feature extraction module according to a preset feature association relationship to obtain a second feature extraction module; the preset associated feature extraction module is used to extract features having a preset feature association relationship with the annotated features from the second sample data after obtaining the annotated features; and combining the first feature extraction module and the second feature extraction module to obtain a feature extraction model, so as to extract the characteristic information from the drug usage data through the feature extraction model.

[0012] According to one aspect of an embodiment of the present application, a medical information query device is provided, including: an information collection module, used to obtain descriptive information of a target object querying drugs or symptoms; an information query module, used to input the descriptive information into a medical information query model to obtain a query result; the medical information query model is obtained by training a preset information query model with first sample data, and the first sample data is obtained by integrating multi-source medical data, and the multi-source medical data includes electronic medical record data, drug sales data, drug testing data, and medication efficacy feedback data.

[0013] According to one aspect of an embodiment of the present application, an electronic device is provided, comprising: one or more processors; a storage device for storing one or more programs, wherein when the one or more programs are executed by the one or more processors, the electronic device implements the medical information query method as described above.

[0014] According to one aspect of an embodiment of the present application, a computer program product is provided, including a computer program, which implements the above-mentioned medical information query method when executed by a processor.

[0015] According to one aspect of an embodiment of the present application, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor of a computer, the computer executes the medical information query method described above.

[0016] Beneficial effects of the present invention: The present invention obtains descriptive information of a target object querying drugs or symptoms, and inputs the descriptive information into a medical information query model to obtain a query result. In the above process, first sample data is obtained by integrating multi-source medical data, and a preset information query model is trained with the first sample data to obtain a medical information query model. The medical information query model integrates multi-source medical data, so that after the descriptive information is input, the output query result is highly correlated with the descriptive information and is more comprehensive, thereby improving the referenceability of the query result.

[0017] It should be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] The drawings herein are incorporated into the specification and constitute a part of the specification, showing embodiments consistent with the present application, and together with the specification, are used to explain the principles of the present application. Obviously, the drawings described below are only some embodiments of the present application, and for those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative work. In the drawings:

[0019] Figure 1 is a schematic diagram of an exemplary system architecture shown in an exemplary embodiment of the present application;

[0020] Figure 2 is a flowchart of a medical information query method shown in an exemplary embodiment of the present application;

[0021] Figure 3 is a schematic diagram of integrating multi-source data according to another exemplary embodiment of the present application;

[0022] Figure 4 A block diagram of a medical information query device suitable for implementing an embodiment of the present application is shown;

[0023] Figure 5 A schematic diagram of the structure of a computer system suitable for implementing an electronic device of an embodiment of the present application is shown. DETAILED DESCRIPTION

[0024] Here, exemplary embodiments will be described in detail, examples of which are shown in the accompanying drawings. When the following description refers to the drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The implementations described in the following exemplary embodiments do not represent all implementations consistent with the present application. Instead, they are only examples of devices and methods consistent with some aspects of the present application as detailed in the attached claims.

[0025] The block diagrams shown in the accompanying drawings are merely functional entities and do not necessarily correspond to physically independent entities. That is, these functional entities may be implemented in software form, or in one or more hardware modules or integrated circuits, or in different networks and / or processor devices and / or microcontroller devices.

[0026] The flowcharts shown in the accompanying drawings are only exemplary and do not necessarily include all the contents and operations / steps, nor must they be executed in the order described. For example, some operations / steps can be decomposed, and some operations / steps can be combined or partially combined, so the actual execution order may change according to actual conditions.

[0027] The term "multiple" as used in this application refers to two or more than two. "And / or" describes the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can mean: A exists alone, A and B exist at the same time, and B exists alone. The character " / " generally indicates that the related objects are in an "or" relationship.

[0028] The technical solution of the embodiment of the present application involves related technologies such as welding of steel structure parts, which is specifically described by the following embodiments:

[0029] Figure 1 It is a schematic diagram of an exemplary system architecture shown in an exemplary embodiment of the present application.

[0030] Reference Figure 1 As shown, the system architecture may include a storage device 101 and a computer device 102. The computer device 102 may be at least one of a desktop graphics processing unit (GPU) computer, a GPU computing cluster, a neural network computer, etc. Relevant technicians may use the computer device 102 to obtain the description information of the target object querying drugs or querying symptoms, input the description information into the medical information query model, and obtain the query result. The storage device 101 is used to store the description information of the target object querying drugs or querying symptoms. In this embodiment, the storage device 101 uses a storage device such as a random access memory to store the description information of the target object querying drugs or querying symptoms, and provides it to the computer device 102 for processing.

[0031] Illustratively, after obtaining the descriptive information of the target object querying drugs or symptoms in the storage device 101, the computer device 102 inputs the descriptive information into the medical information query model to obtain the query result. In the above process, the first sample data is obtained by integrating multi-source medical data, and the preset information query model is trained with the first sample data to obtain the medical information query model. The medical information query model integrates multi-source medical data, so that after the descriptive information is input, the output query result is highly correlated with the descriptive information and is more comprehensive, thereby improving the referenceability of the query result.

[0032] It should be noted that the medical information query method provided in the embodiment of the present application is generally executed by the computer device 102 , and accordingly, the medical information query device is generally set in the computer device 102 .

[0033] The implementation details of the technical solution of the embodiment of the present application are described in detail below:

[0034] Figure 2 is a flowchart of a medical information query method shown in an exemplary embodiment of the present application. The medical information query method can be executed by a computing processing device. The computing processing device can be Figure 1 The computer device 102 shown in FIG. Figure 2 As shown, the medical information query method at least includes steps S210 to S220, which are described in detail as follows:

[0035] In step S210, the description information of the drug or symptom queried by the target object is obtained. In one embodiment of the present application, the target object may be a doctor, a patient, or other person with a use demand. When the target object queries for a drug, the description information may be the drug name, drug number, the function, effect, or dosage of a drug name, etc. When the target object queries for a symptom, the description information may be the symptoms of a certain type of disease, or a text description or voice description of the discomfort symptoms, etc.

[0036] In step S220, the description information is input into the medical information query model to obtain the query result. In one embodiment of the present application, the medical information query model is obtained by training the preset information query model with the first sample data, and the first sample data is obtained by integrating multi-source medical data, and the multi-source medical data includes electronic medical record data, drug sales data, drug testing data, medication efficacy feedback data, and drug description data. The electronic medical record data comes from the electronic medical record system, the drug sales data comes from the drug sales system, the drug testing data comes from the drug experimental platform, and the medication efficacy feedback data comes from the efficacy feedback platform. The first sample data is obtained by integrating multi-source medical data, and the medical information query model is obtained by training the preset information query model with the first sample data. The medical information query model integrates multi-source medical data, so that after the description information is input, the output query result is highly correlated with the description information and is more comprehensive, which improves the referenceability of the query result.

[0037] In another embodiment of the present application, when the target object queries for a drug, the descriptive information may be the drug name, drug number, the function, effect or dosage of a certain drug name, etc. The query results may include not only the function, effect or dosage of the drug name or drug number, but also medication information related to the drug name or drug number (for example, the therapeutic effect achieved by the drug for patients of different ages, genders, symptoms and medication cycles); when the target object queries for a symptom, the query results may include not only the disease symptoms or disease name obtained by the query, but also treatment cases related to the disease symptoms or disease name (for example, the name, dosage and medication cycle of the drug used by patients with the same disease symptoms or disease name).

[0038] In one embodiment of the present application, if the first sample data includes preset subject information and medical information associated with the preset subject information, the process of training the preset information query model through the first sample data to obtain the medical information query model includes:

[0039] The preset subject information is divided into sample subject information and test subject information. In one embodiment of the present application, the preset subject information includes the ID number or drug code of the drug user. When the preset subject information is the ID number of the drug user, the medical information associated with the preset subject information includes: medication cycle, dosage, medication effect, medication method (injection, oral, device medication, etc.). When the preset subject information is the drug code, the medical information associated with the preset subject information includes: drug sales, medication effect, drug description, etc.

[0040] The preset information query model is trained by the sample subject information and the medical information associated with the sample subject information to obtain the trained information query model. In one embodiment of the present application, the preset information query model includes a feature extraction layer, a feature fusion layer and a prediction layer. The feature extraction layer is used to extract features of the sample subject information and the medical information associated with the sample subject information. The feature fusion layer is used to fuse the features extracted from the sample subject information and the features extracted from the medical information associated with the sample subject information, and establish an association relationship between the features extracted from the sample subject information and the features extracted from the medical information associated with the sample subject information. The prediction layer is used to predict based on the fused features and the features with the association relationship to obtain the prediction result; the feature extraction layer, the feature fusion layer and the prediction layer are trained according to the prediction result and the gap between the medical information associated with the sample subject information, and the trained feature extraction layer, the trained feature fusion layer and the trained prediction layer are combined as the trained information query model. The preset information query model can be a neural network model or other models, which are not specifically limited here.

[0041] The test subject information is input into the trained information query model to obtain the test medical information. In one embodiment of the present application, the trained information query model is tested by the test subject information to verify the accuracy of the trained information query model.

[0042] If the error value between the test medical information and the medical information associated with the test subject information is greater than the preset error threshold, the parameters in the trained information query model are adjusted until the error value between the test medical information output by the adjusted information query model and the medical information associated with the test subject information is less than or equal to the preset error threshold, thereby obtaining a medical information query model. In one embodiment of the present application, the preset error threshold can be set according to actual conditions, and when the error value between the test medical information and the medical information associated with the test subject information is greater than the preset error threshold, the parameters in the trained information query model are adjusted, which is beneficial to further improve the accuracy of the trained information query model.

[0043] In one embodiment of the present application, the process of integrating multi-source medical data to obtain first sample data includes:

[0044] Preprocess the electronic medical record data, drug sales data, drug testing data and drug efficacy feedback data to obtain drug use data. In one embodiment of the present application, the preprocessing method includes missing data filling, data deduplication, data denoising, data format conversion and data normalization. The missing data filling method includes mean filling method, median filling method or mode filling method, etc. For example, in the medication cycle data, if the medication cycle of some patients is missing, it can be filled according to the median of the medication cycle of other patients; data deduplication is conducive to removing redundant data in the data and ensuring the uniqueness of the data; data denoising methods include data range checking, data type checking, etc., and by denoising the collected data, it is conducive to removing garbled characters or obviously illogical data (such as age is a negative number) in the collected data, etc., thereby improving the validity and accuracy of the data; data format conversion is conducive to converting the same type of data in the electronic medical record data, drug sales data, drug testing data and drug efficacy feedback data into the same format. For example, there are YYYY in the electronic medical record data. -MM-DD format, while there is a date recorded in MM / DD / YYYY format in the medication efficacy feedback data, the date recorded in MM / DD / YYYY format shall be converted to a date recorded in YYYY-MM-DD format, or the date recorded in YYYY-MM-DD format shall be converted to a date recorded in MM / DD / YYYY format, so as to achieve the unification of data formats; the data normalization methods include numerical data standardization processing methods, categorical data standardization processing methods and text data standardization processing methods, the numerical data include the target object's age, medication dosage, target object's blood glucose test value and blood pressure test value, etc.; the categorical data include the drug number (drug model), the target object's gender, disease type (such as type I or type II diabetes), etc., and the text data include the medication evaluation fed back by the target object, the target object's detailed description of the drug's side effects, the doctor's diagnosis opinion, etc. The processing methods for standardizing categorical data include the one-hot encoding processing method, etc.; the processing methods for standardizing text data can be carried out by establishing a glossary to unify similar words or phrases, for example, "headache" and "headache" are unified as "headache", and at the same time, word vector models and other methods can be used to convert text data into numerical vectors; the processing methods for standardizing numerical data include the standard deviation standardization processing method, the minimum-maximum normalization processing method, etc. By standardizing numerical data, text data and categorical data, it is convenient to integrate and calculate numerical data, text data and categorical data.

[0045] In one embodiment of the present application, the calculation formula of the standard deviation is as follows:

[0046] Z=(X-μ) / σ Formula (1)

[0047] Among them, Z represents the processed data, X is the numerical data, μ is the mean, and σ is the standard deviation.

[0048] By calculating formula (1), the numerical data is converted into a standard normal distribution with a mean of 0 and a standard deviation of 1, which is conducive to eliminating the influence of different units and magnitudes between different data and facilitates large-scale integration and calculation.

[0049] The calculation formula for minimum-maximum normalization is as follows:

[0050] X new =(X-Xmin) / (X max -X min ) Formula (2)

[0051] Among them, X new is the processed data, X is the numerical data, X min is the minimum value of numeric data, X max It is the maximum value of numeric data.

[0052] Feature information is extracted from drug usage data to obtain multiple feature information; multiple feature information is associated according to preset subject information to obtain multiple associated feature information. In one embodiment of the present application, the preset subject information includes an ID number and / or a drug number, and the associated feature information includes feature information of different dimensions, and the feature information of different dimensions includes: ID number, age information, gender information, disease type, residential area, medication cycle, medication dosage, medication effect, medication method (injection, oral, instrument medication, etc.), drug code and drug sales volume, etc. The process of obtaining multiple feature information can be obtained by extracting feature information from drug usage data through a feature extraction model, and the feature extraction model is a combination of a first feature extraction module and a second feature extraction module, the first feature extraction module is obtained by training a preset feature extraction module based on the second sample data, and the second feature extraction module is obtained by training a preset associated feature extraction module based on a preset feature association relationship.

[0053] In another embodiment of the present application, multiple pieces of associated feature information are stored in a matrix form, an array form, etc., which are not specifically limited here. Multiple pieces of associated feature information can be in the order of medication time. The medication time can be the medication time recorded on the efficacy feedback platform, or it can be the medication time estimated based on the drug sales time and medication cycle. It is not specifically limited here. At each medication time point, there is a drug use case. The drug use case includes ID number, age, gender, disease type, residential area, medication dosage, medication cycle, medication effect, etc. Each drug use case is represented by a high-dimensional feature vector, and a feature vector represents a feature information.

[0054] In one embodiment of the present application, multiple pieces of associated feature information are used as first sample data, so that the medical data contained in the first sample data is more comprehensive, and the preset information query model is trained by the first sample data, and the obtained medical information query model can output more comprehensive and rich query results.

[0055] In one embodiment of the present application, the process of integrating multi-source medical data to obtain first sample data includes:

[0056] Preprocess the electronic medical record data, drug sales data, drug testing data and drug efficacy feedback data to obtain drug use data. In one embodiment of the present application, the preprocessing method includes missing data filling, data deduplication, data denoising, data format conversion and data normalization. The missing data filling method includes mean filling method, median filling method or mode filling method, etc. For example, in the medication cycle data, if the medication cycle of some patients is missing, it can be filled according to the median of the medication cycle of other patients; data deduplication is conducive to removing redundant data in the data and ensuring the uniqueness of the data; data denoising methods include data range checking, data type checking, etc., and by denoising the collected data, it is conducive to removing garbled characters or obviously illogical data (such as age is a negative number) in the collected data, etc., thereby improving the validity and accuracy of the data; data format conversion is conducive to converting the same type of data in the electronic medical record data, drug sales data, drug testing data and drug efficacy feedback data into the same format. For example, there are YYYY in the electronic medical record data. -MM-DD format, while there is a date recorded in MM / DD / YYYY format in the medication efficacy feedback data, the date recorded in MM / DD / YYYY format shall be converted to a date recorded in YYYY-MM-DD format, or the date recorded in YYYY-MM-DD format shall be converted to a date recorded in MM / DD / YYYY format, so as to achieve the unification of data formats; the data normalization methods include numerical data standardization processing methods, categorical data standardization processing methods and text data standardization processing methods, the numerical data include the target object's age, medication dosage, target object's blood glucose test value and blood pressure test value, etc.; the categorical data include the drug number (drug model), the target object's gender, disease type (such as type I or type II diabetes), etc., and the text data include the medication evaluation fed back by the target object, the target object's detailed description of the drug's side effects, the doctor's diagnosis opinion, etc. The processing methods for standardizing categorical data include the one-hot encoding processing method, etc.; the processing methods for standardizing text data can be carried out by establishing a glossary to unify similar words or phrases, for example, "headache" and "headache" are unified as "headache", and at the same time, word vector models and other methods can be used to convert text data into numerical vectors; the processing methods for standardizing numerical data include the standard deviation standardization processing method, the minimum-maximum normalization processing method, etc. By standardizing numerical data, text data and categorical data, it is convenient to integrate and calculate numerical data, text data and categorical data.

[0057] In one embodiment of the present application, the calculation formulas for the standard deviation are as shown in formula (1) and formula (2), which will not be described in detail here.

[0058] Feature information is extracted from drug usage data to obtain multiple feature information; multiple feature information is associated according to preset subject information to obtain multiple associated feature information. In one embodiment of the present application, the preset subject information includes an ID number and / or a drug number, and the associated feature information includes feature information of different dimensions, and the feature information of different dimensions includes: ID number, age information, gender information, disease type, residential area, medication cycle, medication dosage, medication effect, medication method (injection, oral, instrument medication, etc.), drug code and drug sales volume, etc. The process of obtaining multiple feature information can be obtained by extracting feature information from drug usage data through a feature extraction model, and the feature extraction model is a combination of a first feature extraction module and a second feature extraction module, the first feature extraction module is obtained by training a preset feature extraction module based on the second sample data, and the second feature extraction module is obtained by training a preset associated feature extraction module based on a preset feature association relationship.

[0059] According to the preset screening dimensions, multiple pieces of associated characteristic information are screened to obtain a screening information set. In one embodiment of the present application, if the preset subject information also includes age information, gender information, disease type, and residential area, the preset screening dimension is at least one of age information, gender information, disease type, and residential area; if the preset subject information also includes sales location and usage method, the preset screening dimension is at least one of sales location and usage method. By setting multiple optional screening dimensions, the flexibility of screening multiple pieces of associated characteristic information is improved.

[0060] The data in the screening information set is used as the first sample data. In one embodiment of the present application, the data in the screening information set is used as the first sample data, so that the first sample data contains fewer dimensions of feature information, and the influence relationship between the feature information can be more intuitively reflected. The preset information query model is trained by the first sample data, and the obtained medical information query model can output more accurate query results.

[0061] In some embodiments of the present application, when the preset screening dimensions are age information and medication cycle, since the age information includes different age stages, for example, 18-40 years old, 41-65 years old, and 66-80 years old, and the medication cycle includes different medication durations, for example, a medication duration of one week, a medication duration of two weeks, and a medication duration of three weeks, when divided according to different age stages and different medication durations, a screening information set of an age stage of 18-40 years old and a medication duration of one week, an age stage of 18-40 years old and a medication duration of two weeks, and an age stage of 18-40 years old and a medication duration of three weeks can be obtained. The above three screening information sets have the same age stage and different medication cycles. Therefore, the above three screening information sets can be used to preliminarily determine the impact of different medication cycles on the medication effect at the age stage of 18-40 years old; it can also be obtained that the age stage is 18-40 years old and the medication duration is one week, the age stage is 41-65 years old and the medication duration is one week, and the age stage is 66 -80 years old and taking medicine for one week. The above three screening information sets have the same medication cycle but different age stages. Therefore, the above three screening information sets can be used to preliminarily determine the influence of different age stages on the medication effect when the medication duration is one week. After preliminarily determining the influence of different age stages on the medication effect when the medication duration is one week, we can further study whether the metabolic levels and underlying diseases (cardiovascular disease, endocrine system disease, digestive system disease, etc.) at different age stages have an impact on the efficacy of the medicine. It can be seen that the data in the screening information set contains fewer dimensions of feature information, which can more intuitively reflect the influence relationship between the feature information of the preset screening dimension and the medication effect. The preset information query model is trained by the data in the screening information set, and the obtained medical information query model can output more accurate query results. Moreover, since there are fewer feature dimensions in the screening information set, the complexity of training the preset information query model is reduced.

[0062] In one embodiment of the present application, the process of filtering multiple pieces of associated feature information according to a preset filtering dimension to obtain a filtered information set includes:

[0063] If the preset subject information also includes age information, gender information, disease type and residential area, at least one of the age information, gender information, disease type and residential area is used as a preset screening dimension to screen multiple pieces of associated feature information to obtain a screening information set. In one embodiment of the present application, the screening information set can be stored in a matrix form, an array form, etc., which is not specifically limited here.

[0064] If the preset subject information also includes the sales location and the usage method, at least one of the sales location and the usage method is used as a preset screening dimension to screen multiple pieces of associated feature information to obtain a screening information set. In one embodiment of the present application, by screening multiple pieces of associated feature information on different preset screening dimensions, the flexibility of screening multiple pieces of associated feature information is improved, which greatly meets the need to train the preset information query model through different screening information sets, and meets the usage needs of the target object to a greater extent.

[0065] In another embodiment of the present application, the developer or researcher selects a screening information set as needed, and preliminarily determines the impact of preset screening dimensions on the medication effect by comparing the data in multiple screening information sets, so as to facilitate the developer or researcher to conduct drug value research and formulate drug optimization strategies, and meet the company's needs for continuous drug optimization and market expansion.

[0066] In one embodiment of the present application, after obtaining the screening information set, the medical information query method further includes:

[0067] If the number of screening information sets is less than the preset number threshold, the data in the screening information set is fitted to obtain a screening feature fitting curve. In one embodiment of the present application, the preset number threshold can be set to 2 or other values. Methods for fitting the data in the screening information set include least squares method, polynomial fitting, etc., which facilitates displaying the data in the screening information set in the form of a fitting curve.

[0068] If the number of screening information sets is greater than or equal to a preset number threshold, the data in each screening information set is fitted to obtain multiple screening feature fitting curves, and the multiple screening feature fitting curves are compared to obtain data differences; if the data difference is greater than or equal to the preset difference value, it is determined that the preset screening dimension has an influence relationship on the medical information associated with the preset subject information; if the data difference is less than the preset difference value, it is determined that the preset screening dimension has no influence relationship on the medical information associated with the preset subject information. In one embodiment of the present application, a fitting curve is used to facilitate observation of the differences in multiple screening feature fitting curves, and the preset difference value is set according to the actual situation. When different age stages and the same medication cycle produce different medication effects (other characteristic information of the comparison subject is the same or there are no other factors affecting the medication effect), and the data difference between different medication effects is greater than or equal to the preset difference value, it means that the age stage has an impact on the medication effect; when different age stages and the same medication cycle (other characteristic information of the comparison subject is the same or there are no other factors affecting the medication effect), the data difference between the medication effects is less than the preset difference value, it means that the age stage has no impact on the medication effect, and the medication effect is characterized by the test data of the patient after medication.

[0069] In one embodiment of the present application, before extracting the characteristic information in the drug usage data, the medical information query method further includes:

[0070] The drug use data within the preset time period is used as the second sample data, and the second sample data is feature-labeled to obtain the labeled data. In one embodiment of the present application, the preset time period can be set according to actual conditions and is not specifically limited here. Methods for feature-labeling the second sample data include annotation methods and tagging methods, etc., which are not specifically limited here.

[0071] The preset feature extraction module is trained by annotating data to obtain a first feature extraction module; and the preset associated feature extraction module is trained according to the preset feature association relationship to obtain a second feature extraction module. In one embodiment of the present application, the preset associated feature extraction module is used to extract features having a preset feature association relationship with the annotated features from the second sample data after obtaining the annotated features; the preset feature extraction module can be a bidirectional encoder representation based on Transformer (Bidirectional Encoder Representations from Transformers, BERT model), or other models, which are not specifically limited here. The preset associated feature extraction module can be a convolutional neural network or a recurrent neural network, etc.

[0072] The first feature extraction module and the second feature extraction module are combined to obtain a feature extraction model, so as to extract feature information from the drug usage data through the feature extraction model. In one embodiment of the present application, after the first feature extraction module extracts feature information from the usage data to obtain the first feature information, it can also extract features having a preset feature association relationship with the first feature information through the second feature extraction module to obtain the second feature information. The first feature information and the second feature information are the feature information finally obtained by the feature extraction model. The first feature extraction module and the second feature extraction module are used in conjunction with each other to perform feature extraction, which makes up for the defect that the extracted feature information is incomplete when only the first feature extraction module is used to perform feature extraction, and improves the accuracy of feature information extraction.

[0073] Figure 3 is a flowchart of integrating multi-source data shown in an exemplary embodiment of the present application, such as Figure 3 As shown, the method for integrating multi-source data includes: (1) data collection: collecting electronic medical record data, drug sales data, drug testing data, and drug efficacy feedback data; (2) data preprocessing: preprocessing the electronic medical record data, drug sales data, drug testing data, and drug efficacy feedback data to obtain drug usage data; (3) feature extraction: extracting feature information from drug usage data to obtain multiple feature information; (4) feature association: associating multiple feature information according to preset subject information to obtain multiple associated feature information.

[0074] The present application realizes multi-source integration of medical data by performing data collection, data preprocessing, feature extraction and feature association, so that the data contained in multiple related feature information is more comprehensive, which can not only meet the needs of training the preset information query model, but also filter multiple related feature information according to the preset filtering dimensions, which is convenient for developers or researchers to select the filtering information set according to their needs, and after selecting the required filtering information set, directly conduct drug value research and formulate drug optimization strategies based on the required filtering information set, saving the time for developers or researchers to screen data, improving the efficiency of developers or researchers in drug value research and formulation of drug optimization strategies, and meeting the needs of developers or researchers for continuous product optimization and market expansion.

[0075] The following describes an embodiment of the device of the present application, which can be used to execute the medical information query method in the above embodiment of the present application. For details not disclosed in the embodiment of the device of the present application, please refer to the embodiment of the medical information query method in the above embodiment of the present application.

[0076] Figure 4 is a block diagram of a medical information query device shown in an exemplary embodiment of the present application. The device can be applied to Figure 1The implementation environment shown in the figure is specifically configured in the computer device 102. The device can also be applied to other exemplary implementation environments and specifically configured in other devices. This embodiment does not limit the implementation environment to which the device is applied.

[0077] like Figure 4 As shown, the exemplary medical information query device includes:

[0078] The information collection module 401 is used to obtain the description information of the drug or disease queried by the target object.

[0079] The information query module 402 is used to input the description information into the medical information query model to obtain the query result.

[0080] In one embodiment of the present application, the target object may be a doctor, a patient, or other person who has a need to use the product. When the target object inquires about a drug, the description information may be the drug name, drug number, the function, effect or dosage of a certain drug name, etc. When the target object inquires about a disease, the description information may be the symptoms of a certain type of disease, or a text description or voice description of the discomfort symptoms, etc.

[0081] In one embodiment of the present application, the medical information query model is obtained by training the preset information query model with the first sample data, and the first sample data is obtained by integrating multi-source medical data, and the multi-source medical data includes electronic medical record data, drug sales data, drug testing data, and medication efficacy feedback data. The electronic medical record data comes from the electronic medical record system, the drug sales data comes from the drug sales system, the drug testing data comes from the drug experimental platform, and the medication efficacy feedback data comes from the efficacy feedback platform. The first sample data is obtained by integrating multi-source medical data, and the medical information query model is obtained by training the preset information query model with the first sample data. The medical information query model integrates multi-source medical data, so that after inputting the description information, the output query result is highly correlated with the description information and is more comprehensive, which improves the referenceability of the query result.

[0082] In another embodiment of the present application, when the target object queries for a drug, the descriptive information may be the drug name, drug number, the function, effect or dosage of a certain drug name, etc. The query results may include not only the function, effect or dosage of the drug name or drug number, but also medication information related to the drug name or drug number (for example, the therapeutic effect achieved by the drug for patients of different ages, genders, symptoms and medication cycles); when the target object queries for a symptom, the query results may include not only the disease symptoms or disease name obtained by the query, but also treatment cases related to the disease symptoms or disease name (for example, the name, dosage and medication cycle of the drug used by patients with the same disease symptoms or disease name).

[0083] It should be noted that the medical information query device provided in the above embodiment and the medical information query method provided in the above embodiment belong to the same concept, wherein the specific manner in which each module and unit performs the operation has been described in detail in the method embodiment and will not be repeated here. In actual application, the medical information query device provided in the above embodiment can distribute the above functions to different functional modules as needed, that is, divide the internal structure of the device into different functional modules to complete all or part of the functions described above, and this is not limited here.

[0084] An embodiment of the present application also provides an electronic device, comprising: one or more processors; a storage device for storing one or more programs, when the one or more programs are executed by one or more processors, the electronic device implements the medical information query method provided in the above-mentioned embodiments.

[0085] Figure 5 The structure diagram of the computer system suitable for implementing the electronic device of the embodiment of the present application is shown. It should be noted that: Figure 5 The computer system 500 of the electronic device shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present application.

[0086] like Figure 5 As shown, the computer system 500 includes a central processing unit (CPU) 501, which can perform various appropriate actions and processes according to the program stored in the read-only memory (ROM) 502 or the program loaded from the storage part 508 to the random access memory (RAM) 503, such as executing the method in the above embodiment. In the RAM 503, various programs and data required for system operation are also stored. The CPU 501, the ROM 502 and the RAM 503 are connected to each other through the bus 504. The input / output (I / O) interface 505 is also connected to the bus 504.

[0087] The following components are connected to the I / O interface 505: an input section 506 including a keyboard, a mouse, etc.; an output section 507 including a cathode ray tube (CRT), a liquid crystal display (LCD), etc., and a speaker; a storage section 508 including a hard disk, etc.; and a communication section 509 including a network interface card such as a LAN (Local Area Network) card, a modem, etc. The communication section 509 performs communication processing via a network such as the Internet. A drive 510 is also connected to the I / O interface 505 as needed. A removable medium 511, such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc., is installed on the drive 510 as needed so that a computer program read therefrom is installed into the storage section 508 as needed.

[0088] In particular, according to an embodiment of the present application, the process described above with reference to the flowchart can be implemented as a computer software program. For example, an embodiment of the present application includes a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program includes a computer program for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from a network through a communication section 509, and / or installed from a removable medium 511. When the computer program is executed by a central processing unit (CPU) 501, various functions defined in the system of the present application are executed.

[0089] It should be noted that the computer-readable medium shown in the embodiment of the present application can be a computer-readable signal medium or a computer-readable storage medium or any combination of the above two. The computer-readable storage medium can be, for example, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples of computer-readable storage media can include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM), a flash memory, an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present application, a computer-readable signal medium may include a data signal propagated in a baseband or as part of a carrier wave, wherein a computer-readable computer program is carried. This propagated data signal can take a variety of forms, including but not limited to an electromagnetic signal, an optical signal, or any suitable combination of the above. A computer-readable signal medium may also be any computer-readable medium other than a computer-readable storage medium, which may send, propagate or transmit a program for use by or in conjunction with an instruction execution system, apparatus or device. A computer program contained on a computer-readable medium may be transmitted using any appropriate medium, including but not limited to: wireless, wired, etc., or any suitable combination of the above.

[0090] The flowchart and block diagram in the accompanying drawings illustrate the possible architecture, functions and operations of the system, method and computer program product according to various embodiments of the present application. Wherein, each box in the flowchart or block diagram can represent a module, a program segment, or a part of the code, and the above-mentioned module, program segment, or a part of the code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order from the order marked in the accompanying drawings. For example, two boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram or flowchart, and the combination of boxes in the block diagram or flowchart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0091] The units involved in the embodiments described in this application may be implemented by software or hardware, and the units described may also be set in a processor. The names of these units do not, in some cases, constitute limitations on the units themselves.

[0092] Another aspect of the present application also provides a computer program product, including a computer program, which implements the medical information query method provided in the above-mentioned embodiments when the computer program is executed by a processor.

[0093] Another aspect of the present application further provides a computer-readable storage medium on which computer-readable instructions are stored. When the computer-readable instructions are executed by a processor of a computer, the computer executes the medical information query method provided in the above-mentioned embodiments. The computer-readable storage medium may be included in the electronic device described in the above-mentioned embodiments, or may exist independently without being assembled into the electronic device.

[0094] It should be noted that, although several modules or units of the equipment for action execution are mentioned in the above detailed description, this division is not mandatory. In fact, according to the embodiments of the present application, the features and functions of two or more modules or units described above can be embodied in one module or unit. On the contrary, the features and functions of one module or unit described above can be further divided into being embodied by multiple modules or units.

[0095] Through the description of the above implementation methods, it is easy for those skilled in the art to understand that the example implementation methods described here can be implemented by software or by combining software with necessary hardware. Therefore, the technical solution according to the implementation methods of the present application can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (which can be a CD-ROM, a USB flash drive, a mobile hard disk, etc.) or on a network, and includes several instructions to enable a computing device (which can be a personal computer, a server, a touch terminal, or a network device, etc.) to execute the method according to the implementation methods of the present application.

[0096] Those skilled in the art will readily appreciate other embodiments of the present application after considering the specification and practicing the embodiments disclosed herein. The present application is intended to cover any variations, uses or adaptations of the present application, which follow the general principles of the present application and include common knowledge or customary technical means in the art that are not disclosed in the present application.

[0097] It should be understood that the above content is only a preferred exemplary embodiment of the present application and is not intended to limit the implementation scheme of the present application. A person skilled in the art can easily make corresponding changes or modifications based on the main concept and spirit of the present application. Therefore, the scope of protection of the present application shall be based on the scope of protection required by the claims.

Claims

1. A medical information query method, characterized in that: include: Obtain the description information of the target object's query for drugs or symptoms; Inputting the description information into a medical information query model to obtain a query result; The medical information query model is obtained by training a preset information query model through first sample data, and the first sample data is obtained by integrating multi-source medical data, and the multi-source medical data includes electronic medical record data, drug sales data, drug testing data, and medication efficacy feedback data.

2. The medical information query method according to claim 1, characterized in that: If the first sample data includes preset subject information and medical information associated with the preset subject information, the process of training the preset information query model through the first sample data to obtain the medical information query model includes: Dividing the preset subject information into sample subject information and test subject information; The preset information query model is trained by using the sample subject information and the medical information associated with the sample subject information to obtain a trained information query model; Inputting the test subject information into the trained information query model to obtain test medical information; If the error value between the test medical information and the medical information associated with the test subject information is greater than a preset error threshold, the parameters in the trained information query model are adjusted until the error value between the test medical information output by the adjusted information query model and the medical information associated with the test subject information is less than or equal to the preset error threshold, thereby obtaining the medical information query model.

3. The medical information query method according to claim 1, characterized in that: The process of integrating multi-source medical data to obtain the first sample data includes: Preprocessing the electronic medical record data, the drug sales data, the drug testing data and the drug efficacy feedback data to obtain drug usage data; the preprocessing method includes missing data filling, data deduplication, data denoising, data format conversion and data normalization; Extracting characteristic information from the drug usage data to obtain a plurality of characteristic information; associating the plurality of characteristic information according to preset subject information to obtain a plurality of associated characteristic information; the preset subject information includes an identity card number and / or a drug number; The plurality of pieces of associated feature information are used as the first sample data.

4. The medical information query method according to claim 1, characterized in that: The process of integrating multi-source medical data to obtain the first sample data includes: Preprocessing the electronic medical record data, the drug sales data, the drug testing data and the drug efficacy feedback data to obtain drug usage data; the preprocessing method includes missing data filling, data deduplication, data denoising, data format conversion and data normalization; Extracting characteristic information from the drug usage data to obtain a plurality of characteristic information; associating the plurality of characteristic information according to preset subject information to obtain a plurality of associated characteristic information; the preset subject information includes an identity card number and / or a drug number; Filtering the plurality of pieces of associated feature information according to a preset filtering dimension to obtain a filtered information set; The data in the screening information set is used as the first sample data.

5. The medical information query method according to claim 4, characterized in that: The process of filtering the plurality of associated feature information according to the preset filtering dimension to obtain the filtered information set includes: If the preset subject information further includes age information, gender information, disease type and residential area, at least one of the age information, gender information, disease type and residential area is used as the preset screening dimension to screen the multiple pieces of associated feature information to obtain the screening information set; If the preset subject information also includes a sales location and a usage method, at least one of the sales location and the usage method is used as the preset screening dimension to screen the multiple pieces of the associated feature information to obtain the screening information set.

6. The medical information query method according to claim 4 or 5, characterized in that: After obtaining the screening information set, the method further includes: If the number of the screening information sets is less than a preset number threshold, fitting the data in the screening information sets to obtain a screening feature fitting curve; If the number of the screening information sets is greater than or equal to a preset number threshold, the data in each screening information set is fitted to obtain multiple screening feature fitting curves, and the multiple screening feature fitting curves are compared to obtain data differences; if the data difference is greater than or equal to the preset difference value, it is determined that the preset screening dimension has an influence relationship on the medical information associated with the preset subject information; if the data difference is less than the preset difference value, it is determined that the preset screening dimension has no influence relationship on the medical information associated with the preset subject information.

7. The medical information query method according to any one of claims 3 to 5, characterized in that: Before extracting the characteristic information from the drug usage data, the method further includes: Using the drug usage data within a preset time period as second sample data, and performing feature annotation on the second sample data to obtain annotated data; The preset feature extraction module is trained by the annotated data to obtain a first feature extraction module; and the preset associated feature extraction module is trained according to the preset feature association relationship to obtain a second feature extraction module; the preset associated feature extraction module is used to extract features having a preset feature association relationship with the annotated features from the second sample data after obtaining the annotated features; The first feature extraction module and the second feature extraction module are combined to obtain a feature extraction model, so as to extract feature information from the drug usage data through the feature extraction model.

8. A medical information query device, characterized in that: include: The information collection module is used to obtain the description information of the target object's query of drugs or symptoms; An information query module, used for inputting the description information into a medical information query model to obtain a query result; The medical information query model is obtained by training a preset information query model through first sample data, and the first sample data is obtained by integrating multi-source medical data, and the multi-source medical data includes electronic medical record data, drug sales data, drug testing data, and medication efficacy feedback data.

9. An electronic device, characterized in that: include: one or more processors; A storage device for storing one or more programs, which, when executed by the one or more processors, enables the electronic device to implement the medical information query method as described in any one of claims 1 to 7.

10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the medical information query method as described in any one of claims 1 to 7 is implemented.

Images