Medical data prediction method, device, computer equipment and storage medium

By acquiring and processing medical data, generating prediction historical data and inputting prediction models, the problem of inefficient prediction of traditional medical data is solved and efficient medical data prediction is achieved.

CN109543904BActive Publication Date: 2024-11-05SHENZHEN PING AN MEDICAL HEALTH TECHNOLOGY SERVICES CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN201811401537.3
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Priority Date
2018-05-29
Filing Date
2018-11-22
Publication Date
2024-11-05
Estimated Expiration
2038-11-22

AI Technical Summary

Technical Problem

Traditional medical data prediction methods require manual analysis of large amounts of data, resulting in inefficiency.

Method used

By obtaining the medical data to be predicted and historical medical data, selecting the corresponding prediction methods, performing standardized processing and parameter processing, generating predicted historical medical data, and inputting the prediction model to obtain prediction results, reducing manual processing.

Benefits of technology

It improves the efficiency of medical data prediction, reduces the need for manual processing of data, and improves prediction efficiency and accuracy.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN109543904B_ABST
    Figure CN109543904B_ABST
Patent Text Reader

Abstract

The present application relates to a medical data prediction method, device, computer equipment and storage medium. The method includes: obtaining medical data to be predicted and historical medical data corresponding to the medical data to be predicted, selecting a prediction method corresponding to the medical data to be predicted; standardizing the historical medical data to obtain standardized historical medical data; according to the prediction method, processing the prediction parameters in the standardized historical medical data to obtain predicted historical medical data; inputting the predicted historical medical data into a prediction model, obtaining a prediction result corresponding to the medical data to be predicted according to the prediction method, and outputting the prediction result, and the prediction model is used to characterize the corresponding relationship between the predicted historical medical data and the prediction result. The use of this method can improve the prediction efficiency without the need for manual processing of a large amount of data.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application relates to the field of computer technology, and in particular to a medical data prediction method, apparatus, computer device and storage medium. Background Art

[0002] With the progress and development of society, it is crucial to understand the medical data of different users and predict the medical behavior of future users.

[0003] Traditionally, when using medical data to predict future medical behavior, manual analysis of the collected data, such as the corresponding medical expenses, is often required. This results in a large workload for manual analysis and leads to low efficiency. Summary of the invention

[0004] Based on this, it is necessary to provide a medical data prediction method, device, computer equipment and storage medium that can improve analysis efficiency in response to the above technical problems.

[0005] A medical data prediction method, the method comprising:

[0006] Acquire medical data to be predicted and historical medical data corresponding to the medical data to be predicted, and select a prediction method corresponding to the medical data to be predicted;

[0007] Standardizing the historical medical data to obtain standardized historical medical data;

[0008] According to the prediction method, the prediction parameters in the standardized historical medical data are processed to obtain predicted historical medical data;

[0009] The predicted historical medical data is input into the prediction model, a prediction result corresponding to the medical data to be predicted is obtained according to the prediction method, and the prediction result is output. The prediction model is used to characterize the corresponding relationship between the predicted historical medical data and the prediction result.

[0010] In one embodiment, the step of processing the prediction parameters in the standardized historical medical data to obtain the predicted historical medical data according to the prediction method includes:

[0011] According to the prediction method, querying the region information associated with the medical data to be predicted;

[0012] Selecting discarded data from the standardized historical medical data according to the regional information;

[0013] Adding a deprecated label to the deprecated data;

[0014] The standardized historical medical data without the abandoned labels is extracted to obtain predicted historical medical data.

[0015] In one embodiment, the method further comprises:

[0016] When the prediction method is selected as a linear prediction method, querying whether there is data to be replaced in the standardized historical data;

[0017] When there is data to be replaced, querying a first preset replacement value corresponding to the data to be replaced;

[0018] The data to be replaced is replaced with the first preset replacement value to obtain predicted historical medical data.

[0019] In one embodiment, after querying the first preset replacement data value corresponding to the data to be replaced, the method further includes:

[0020] When the first preset replacement value corresponding to the data to be replaced is not found, then querying the replacement range corresponding to the data to be replaced;

[0021] A second preset replacement value corresponding to the replacement range is selected, and the data to be replaced is replaced with the second preset replacement value to obtain predicted historical medical data.

[0022] In one embodiment, the step of standardizing the historical medical data to obtain standardized historical medical data includes:

[0023] Acquire the data to be processed in the standardized historical medical data, and extract the data fields corresponding to the data to be processed;

[0024] Selecting conversion logic corresponding to the data field;

[0025] According to the conversion logic, the data to be processed of the corresponding field is converted into standardized historical medical data.

[0026] In one embodiment, the step of inputting the predicted historical medical data into a prediction model, obtaining a prediction result corresponding to the medical data to be predicted according to the prediction method, and outputting the prediction result comprises:

[0027] Acquire the medical data to be predicted whose cost exceeds a preset value in the prediction result;

[0028] Obtaining identity information corresponding to the medical data to be predicted whose cost exceeds a preset value;

[0029] The medical behavior corresponding to the identity information is monitored.

[0030] A medical data prediction device, comprising:

[0031] An acquisition module, used to acquire the medical data to be predicted and the historical medical data corresponding to the medical data to be predicted, and select a prediction method corresponding to the medical data to be predicted;

[0032] A historical medical data processing module, used for standardizing the historical medical data to obtain standardized historical medical data;

[0033] A prediction parameter processing module, used for processing the prediction parameters in the standardized historical medical data according to the prediction method to obtain predicted historical medical data;

[0034] The prediction module is used to input the predicted historical medical data into the prediction model, obtain the prediction result corresponding to the medical data to be predicted according to the prediction method, and output the prediction result. The prediction model is used to characterize the corresponding relationship between the predicted historical medical data and the prediction result.

[0035] In one embodiment, the prediction parameter processing module includes:

[0036] A query unit, used for querying the region information associated with the medical data to be predicted according to the prediction method;

[0037] A selection unit, configured to select discarded data from the standardized historical medical data according to the regional information;

[0038] An adding unit, used for adding a discarded label to the discarded data;

[0039] An extraction unit is used to extract the standardized historical medical data without the abandoned label to obtain predicted historical medical data.

[0040] A computer device comprises a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps of the above method when executing the computer program.

[0041] A computer-readable storage medium stores a computer program, which implements the steps of the above method when executed by a processor.

[0042] The above-mentioned medical data prediction method, device, computer equipment and storage medium do not require manual processing of large amounts of data. They can select corresponding prediction methods according to the medical data to be predicted, and process historical medical data to obtain predicted historical medical data, thereby using predicted historical medical data to predict the medical data to be predicted, thereby improving prediction efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] Figure 1 This is an application scenario diagram of a medical data prediction method in one embodiment;

[0044] Figure 2 is a flow chart of a medical data prediction method in one embodiment;

[0045] Figure 3 A schematic diagram of a process flow of a prediction parameter processing step in one embodiment;

[0046] Figure 4 is a structural block diagram of a medical data prediction device in one embodiment;

[0047] Figure 5 FIG. 4 is a diagram showing the internal structure of a computer device in one embodiment. DETAILED DESCRIPTION

[0048] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0049] The medical data prediction method provided in this application can be applied to Figure 1 In the application environment shown. Among them, the prediction platform 102 communicates with the server 104 through the network. When the prediction platform 102 obtains the medical data to be predicted, it obtains the historical medical data corresponding to the medical data to be predicted from the server 104, and then the prediction platform selects the prediction method corresponding to the medical data to be predicted. The prediction platform 102 standardizes the historical medical data obtained from the server 104 to obtain standardized historical medical data. The prediction platform 102 processes the prediction parameters in the obtained standardized historical medical data according to the selected prediction method to obtain predicted historical medical data, and then the prediction platform 102 obtains the corresponding prediction model from the server, inputs the predicted historical medical data into the prediction model to obtain the corresponding prediction result, and outputs the prediction result. Among them, the prediction platform 102 can be implemented with an independent server or a server cluster composed of multiple servers, and the server 104 can be implemented with an independent server or a server cluster composed of multiple servers.

[0050] In one embodiment, Figure 2 As shown, a medical data prediction method is provided, which is applied to Figure 1 The prediction platform in is taken as an example to illustrate, including the following steps:

[0051] S202: Obtain medical data to be predicted and historical medical data corresponding to the medical data to be predicted, and select a prediction method corresponding to the medical data to be predicted.

[0052] Specifically, the medical data to be predicted refers to the medical behavior data of the user that can be predicted within a preset time period. The medical data to be predicted may be cost data, diagnosis data, or prescription data, etc. The data to be predicted contains corresponding identity information. For example, the medical data to be predicted may be the cost required for the user to see a doctor in the next year, the number of hospitalizations of the user in the next year, the number of medicines purchased by the user in the next year, etc. The historical medical data refers to the medical behavior data of the user that already exists within the preset time period. The historical medical data may be at least one of cost data, spent cost data, existing diagnosis data, and existing prescription data, and the historical medical data contains identity information consistent with the corresponding data to be predicted. For example, the historical medical data may be at least one of the cost required for the user to see a doctor in the historical preset time period, the number of hospitalizations of the user in the historical preset time period, the number of outpatient treatments of the user in the historical preset event period, and the number of medicines purchased by the user in the historical preset time period. The prediction method can be selected according to the type of medical data to be predicted. For example, when the medical data to be predicted is displayed as continuous linear data, the prediction method can be selected as a linear prediction method. When the data to be predicted is displayed as Boolean data, the prediction method can be selected as a Boolean prediction method. For example, when the medical data to be tested is the cost of medical treatment, the cost of medical treatment can be a continuous linear variable, and the prediction method can be selected as a linear prediction method. When the medical data to be predicted is whether the patient will be hospitalized within a preset time period, the prediction result output according to the historical medical data is hospitalization or not. The data to be predicted on whether the patient will be hospitalized is Boolean prediction data, and the prediction method to be selected is a Boolean prediction method.

[0053] Specifically, when the prediction platform obtains the medical data to be predicted, the prediction platform extracts the identity information contained in the data to be predicted, and then queries the historical medical data corresponding to the identity information of the data to be predicted based on the identity information contained in the data to be predicted, and then selects the prediction method corresponding to the data to be predicted based on the data to be predicted. For example, if the prediction platform obtains the medical data to be predicted to predict the hospitalization expenses of user A in the next year, then based on the identity information of user A carried by the data to be predicted, such as name, age, gender, ID card information, etc., the corresponding medical information of the user in the past two years is obtained, such as the corresponding disease information, number of hospitalizations and hospitalization expenses of user A, and then based on the fact that the measured data is the hospitalization expenses in the next year, which is continuous linear data, the selected prediction method is the linear prediction method.

[0054] S204: Standardizing the historical medical data to obtain standardized historical medical data.

[0055] Specifically, standardized historical data refers to the standardized processing of the acquired historical medical data according to the corresponding fields, so that the historical medical data of the same category is displayed in a standardized format, and the standardized historical medical data can be used for subsequent predictions. The standardized historical data can be at least one of the standardized cost data, the standardized medical data, and the standardized prescription information. Specifically, when the prediction platform acquires the historical medical data, the data category of the historical medical data is queried, and the historical medical data is standardized according to the queried data category, so as to obtain standardized historical data. For example, the prediction platform acquires the historical medical data, and the data category of the historical medical data is queried as cost data, and then the cost data is refined, that is, refined into hospitalization expenses, prescription costs or hospitalization expenses, etc., so that the refined cost data is standardized historical medical data.

[0056] S206: According to the prediction method, the prediction parameters in the standardized historical medical data are processed to obtain predicted historical medical data.

[0057] Specifically, prediction parameters refer to parameters that are included in standardized historical data and need to be involved in prediction, and the prediction parameters can be processed, such as replacing the prediction parameters that need to be replaced to obtain parameters that can be directly used. Predicted historical medical data refers to medical behavior data that contains parameters that can directly participate in prediction, that is, predicted historical medical data refers to processing the parameters contained in the standardized historical medical data that are not conducive to direct application for prediction, so that the standardized historical data can be directly used for prediction. Specifically, the prediction platform is based on the medical data to be predicted, and the historical medical data that can be used to predict the medical data to be predicted, and standardizes the historical medical data to obtain standardized historical medical data. The prediction platform selects the corresponding prediction method based on the medical data to be predicted, and replaces the prediction parameters that cannot be directly used in the standardized historical medical data according to the prediction method to obtain predicted medical data, so that the predicted medical data can be directly used for prediction.

[0058] S208: Input the predicted historical medical data into the prediction model, obtain the prediction result corresponding to the medical data to be predicted according to the prediction method, and output the prediction result. The prediction model is used to characterize the corresponding relationship between the predicted historical medical data and the prediction result.

[0059] Specifically, the prediction model refers to a model that can characterize the corresponding results between the predicted historical medical data and the predicted results, that is, by using the prediction model, the predicted results of the medical data to be predicted can be obtained according to the predicted historical medical data, and the prediction model can predict continuous linear data and Boolean data. Specifically, when predicting the medical data to be predicted, the predicted historical medical data to be processed is input into the prediction model, and then the prediction model queries the corresponding prediction method, and predicts the prediction results according to the prediction parameters in the input historical medical data, and outputs the prediction results corresponding to the data to be predicted. For example, if the medical data to be predicted is the number of hospitalizations of user A in the next year, the linear prediction method is selected as the prediction method, and the disease data in the historical medical data corresponding to user A is replaced with the standardized historical medical data obtained by the International Classification of Diseases. In addition, the prediction parameters in the standardized historical medical data can be replaced to obtain predicted historical medical data. For example, the number of hospitalizations already included can be replaced with the hospitalization expenses, which facilitates the use of linear prediction method for prediction. The prediction platform inputs the predicted historical medical data into the prediction model. The prediction model queries the prediction results corresponding to the corresponding predicted historical medical data based on the input predicted historical medical data and the linear prediction method, and then outputs the prediction results.

[0060] It should be noted that the prediction model can be obtained by training with corresponding training sample data, which can be obtained by obtaining sample data and dividing the sample data into training data and test data, standardizing the training sample data to obtain standardized training sample data, taking the training sample data as feature values, taking the known prediction results corresponding to the training sample data, and then using the sample data as feature values ​​and the prediction results to calculate the prediction relationship, and then using the test data and the calculated prediction relationship to query the corresponding test prediction result corresponding to the test data. When the test prediction result is a target test prediction result, the calculated prediction relationship can be applied, that is, the prediction relationship can be used for prediction. Model, the prediction model can be used to verify the calculated prediction relationship. When the verification is successful, the prediction model can be used for prediction. During the training process of the prediction model, the corresponding linear regression algorithm can be used to calculate the training data as the feature quantity and the prediction results corresponding to the training data using the linear regression algorithm to obtain the corresponding prediction relationship. After verification, a prediction model that can be used for prediction is obtained. When training the training model, the logistic regression algorithm can also be used to calculate the corresponding prediction relationship, or the random forest algorithm can be used to calculate the prediction relationship. Therefore, the obtained prediction model can predict both continuous linear data and Boolean data.

[0061] In this embodiment, there is no need to manually process a large amount of data. A corresponding prediction method can be selected according to the medical data to be predicted, and the historical medical data can be standardized to obtain standardized historical medical data. The prediction parameters in the standardized historical medical data can be processed to obtain predicted medical data, and then the predicted medical data can be used to predict the medical data to be predicted, thereby improving the prediction efficiency.

[0062] In one embodiment, see Figure 3 , a flowchart of a prediction parameter processing step is provided, the prediction parameter processing step, that is, according to the prediction method, the prediction parameters in the standardized historical medical data are processed to obtain predicted historical medical data, including: according to the prediction method, querying the regional information associated with the medical data to be predicted; according to the regional information, selecting abandoned data in the standardized historical medical data; adding abandoned labels to the abandoned data; extracting the standardized historical medical data without the abandoned labels to obtain predicted historical medical data.

[0063] Specifically, the regional information refers to the region corresponding to the identity information contained in the medical data to be predicted, that is, it can be the source region of the user associated with the identity information contained in the medical data to be predicted. Abandoned data refers to the discarded parameters contained in the standardized historical medical data in the process of using the model for prediction, that is, the feature values ​​that are not considered in the prediction process. Abandoned labels refer to the identification of abandoned data. When there is abandoned data in the standardized medical data, the corresponding abandoned labels are added. When the prediction model is used for prediction, the data with the abandoned labels added will not participate in the prediction.

[0064] Specifically, when predicting the medical data to be predicted, it is necessary to use standardized historical medical data for prediction, and some data contained in the standardized historical medical data may be accidental data. In the process of prediction, considering the accidental data may affect the accuracy of the prediction, so it is necessary to discard the accidental data. The prediction platform obtains the medical data to be predicted, and obtains the historical medical data corresponding to the selected medical data to be predicted, and standardizes the historical medical data to obtain the standardized historical medical data. The prediction platform queries the identity information contained in the medical data to be predicted, queries the regional information corresponding to the identity information, and queries the pre-stored data corresponding to the regional information according to the regional information. It queries whether the standard historical medical data contains data that does not need to be involved in the prediction. When it contains data that does not need to be involved in the prediction, the data that does not need to be involved in the prediction in the standard historical medical data is marked as abandoned data, thereby adding an abandoned label to the abandoned data, and then the abandoned data is not used in the subsequent prediction process, so that the prediction platform extracts the standardized historical medical data without the abandoned label as the predicted historical medical data, and then inputs the historical predicted medical data into the corresponding prediction model for prediction.

[0065] For example, it is preset that in region R, disease data a is accidental data. When the prediction platform receives the medical data to be predicted to predict the hospitalization expenses of user A in the next year, it obtains the historical medical data corresponding to user A, and then standardizes the historical medical data to obtain standardized historical data. The prediction platform queries user A and finds that the region to which user A belongs is region R, and the accidental data pre-stored in region R is disease data a. It queries whether the standardized historical data contains disease data a. When disease data a is contained, if disease data a is involved in the prediction, since disease data a appears as accidental data, if disease data a is considered in the data prediction, the predicted hospitalization expenses will be biased. Therefore, disease data a is added with a deprecated label, and disease data a will not be involved in the subsequent prediction. The prediction platform extracts other data without deprecated labels as predicted medical data.

[0066] In this embodiment, the acquired historical medical data is standardized to obtain standardized historical medical data, and then the standardized medical data will discard any accidental data that appears, that is, add corresponding discarded labels to avoid the appearance of accidental data and cause inaccurate results in predicting the medical data to be predicted, thereby improving the prediction accuracy.

[0067] In one embodiment, see Figure 3The medical data prediction method may also include: when the prediction method selected is the linear prediction method, querying whether there is data to be replaced in the standardized historical data; when there is data to be replaced, querying the first preset replacement value corresponding to the data to be replaced; and replacing the data to be replaced with the first preset replacement value to obtain the predicted historical medical data.

[0068] Specifically, the data to be replaced refers to the standardized historical data that needs to be replaced when the prediction is made according to the prediction method because the selected prediction method cannot directly use the standardized historical data. For example, when the selected prediction method is the linear prediction method, when the standardized historical data contains the number of hospitalizations, and the number of hospitalizations may jump, participating in the prediction will lead to inaccurate prediction results, and the linear method cannot be used directly for prediction, so the number of hospitalizations needs to be replaced. The first preset replacement value refers to the data that can be replaced with the data that can be applied to the prediction model and the data that is predicted according to the prediction method. For example, the medical data to be predicted is the number of hospitalizations, which may be jump data. When the jump data is used directly for prediction, it will result in a large error. In this case, the average cost corresponding to the number of hospitalizations can be used for replacement, and the average cost of replacing the number of hospitalizations is the first preset replacement data.

[0069] Specifically, when the prediction platform obtains the medical data to be predicted and the prediction method is a linear prediction method, it further obtains historical medical data, and standardizes the historical medical data to obtain standardized historical medical data, obtains preset data to be replaced, and compares the preset data to be replaced with the standardized historical medical data. When the comparison is successful, the standardized medical data contains data to be replaced, and then the prediction platform queries the first preset replacement data corresponding to the data to be replaced, and replaces the data to be replaced with the first preset replacement data, thereby obtaining predicted historical medical data. Furthermore, when selecting the first preset replacement data, in order to ensure subsequent accuracy, regional differences are considered. Therefore, when the prediction platform queries the data to be replaced, it can obtain the identity information contained in the medical data to be predicted, and then query the associated regional information based on the identity information, thereby querying the first preset replacement value corresponding to the regional information, and then using the first preset replacement value corresponding to the region to replace the data to be replaced to obtain predicted historical medical data.

[0070] For example, when the prediction platform obtains the medical data to be predicted, which is the number of hospitalizations of user A in the next year, it obtains the historical medical data corresponding to user A, such as user A's disease data a for two years and the number of hospitalizations being 3 times, and then standardizes the historical medical data to obtain standardized historical medical data. The prediction platform obtains the preset data to be replaced, which is the number of hospitalizations. Taking into account regional differences, the identity information of user A is obtained, and the regional information of user A is queried as region R. The average cost of different number of hospitalizations in region R corresponding to the disease data of user A is obtained, that is, the average cost of three hospitalizations due to disease a in region R is obtained, and the average cost is used as the first preset replacement data, and the average cost is used to replace the number of hospitalizations to obtain the predicted historical medical data.

[0071] In this embodiment, since there may be data to be replaced in the standardized historical data, such as jump data, which will lead to inaccurate prediction results when making predictions, the data to be replaced is replaced with a first preset replacement value to obtain predicted historical medical data, thereby ensuring the accuracy of the prediction.

[0072] In one embodiment, see Figure 3 After querying the first preset replacement data value corresponding to the data to be replaced, it also includes: when the first preset replacement value corresponding to the data to be replaced is not queried, querying the replacement range corresponding to the data to be replaced; selecting the second preset replacement value corresponding to the replacement range, and using the second preset replacement value to replace the data to be replaced to obtain predicted historical medical data.

[0073] Specifically, the replacement range refers to the range containing the data to be replaced, and different replacement ranges correspond to different preset data, so that the data to be replaced within the same replacement range can be replaced with the same preset data. The second predicted replacement value refers to the preset data corresponding to the replacement range, and the second predicted replacement value can replace the data to be replaced within the replacement range, thereby obtaining data that can participate in the prediction according to the prediction method, thereby predicting the medical data to be predicted. For example, the medical data to be predicted is the number of prescriptions in the prescription information. Since the number of prescriptions may be jump data, and the number of prescriptions may have multiple times, if each number of prescriptions is replaced with corresponding preset data, the amount of data is too large, and data jumps cannot be avoided. Therefore, the replacement range corresponding to the number of prescriptions can be queried, so as to query the average cost corresponding to the replacement range for replacement, so that the average cost corresponding to the replacement range of the number of prescriptions is replaced as the second preset replacement data.

[0074] Specifically, when the preset platform queries the data to be replaced in the standardized historical medical data, it queries the first preset replacement value of the data to be replaced. When the first preset replacement value is not queried, the replacement range corresponding to the data to be replaced is obtained, and based on the queried replacement range, the second preset replacement value corresponding to the replacement range is queried, so that the second preset replacement value is used to replace the data to be replaced. Furthermore, when selecting the second preset replacement value, in order to ensure subsequent accuracy, regional differences are taken into consideration. When the prediction platform queries the data to be replaced, the identity information contained in the medical data to be predicted is obtained, and then the associated regional information is queried based on the identity information, the replacement range corresponding to the regional information is queried, and then the second preset replacement value corresponding to the replacement range is queried, so that the second preset replacement value is used to replace the replacement data to obtain predicted historical medical data.

[0075] For example, when the prediction platform obtains the medical data to be predicted, which is the number of prescriptions for user B in the next year, the historical medical data corresponding to user B is obtained, such as the disease data b of user B in the past two years, the number of prescriptions is 13 times, etc., and then the number of prescriptions is queried as the data to be replaced, but the first preset replacement value is not directly queried, then the replacement range corresponding to the number of prescriptions is queried, that is, the identity information of user B is obtained from the prediction platform, and the regional information of user B is queried as region S, and the replacement range of the number of prescriptions corresponding to the disease data of user B in region S is obtained, so as to query the replacement range corresponding to the number of prescriptions of 13 times, such as the replacement range queried is 10 to 20 times, so as to query the second preset replacement value corresponding to the replacement range, that is, the unified average cost of prescriptions corresponding to 10 to 20 times is queried as F, so as to use the unified average cost of prescriptions F corresponding to 10 to 20 times as the second preset replacement value, and replace the data to be replaced to obtain the predicted historical medical data.

[0076] In this embodiment, when the first preset replacement value is not found, the data to be replaced may be too much, so the second preset replacement value corresponding to the replacement range is used for replacement, thereby ensuring accurate predicted historical medical data and making the prediction result accurate.

[0077] In one of the embodiments, historical medical data is standardized to obtain standardized historical medical data, including: obtaining the data to be processed in the standardized historical medical data, and extracting the data field corresponding to the data to be processed; selecting the conversion logic corresponding to the data field; and converting the data to be processed of the corresponding field according to the conversion logic to obtain the standardized historical medical data.

[0078] Specifically, the data to be processed refers to the data contained in the standardized historical data that is not displayed in a standard format or display mode, and since the data to be processed is not displayed in a standard format or display mode, it is necessary to perform standardized processing for use when using the prediction model for prediction. The data field refers to the field in the data to be processed that displays the data category of the data to be processed. For example, the data field can identify the field to be processed as cost data, disease data, drug data, diagnosis data, etc. The conversion logic refers to the conversion rule that can convert the data to be processed into data that can be used for prediction. The conversion rule can be to display the data to be processed in a standard format or display mode. For example, the conversion logic can be to indicate that the data to be processed that is the cost is split to obtain detailed costs, such as the overall cost is split into detailed costs, and hospitalization costs, outpatient costs, prescription costs, etc. are obtained respectively; the conversion logic can indicate that the disease data is represented by the International Classification of Diseases code; the conversion logic can be to indicate that the hospitalization date is replaced by the number of hospitalizations; the conversion logic can replace the visit date with the number of visits, etc.

[0079] Specifically, when the prediction platform obtains historical medical data, it queries whether there is data to be processed in the historical medical data according to the pre-stored data to be processed. When there is data to be processed, the field of the data to be processed is obtained, and the category of the data to be processed is obtained according to the field. Then, according to the field of the data to be processed, the conversion logic applicable to the data to be processed is obtained. According to the conversion logic, the data to be processed is converted according to the conversion logic, thereby obtaining standardized historical medical data. For example, when the prediction platform obtains historical medical data, the pre-stored data to be processed is cost data. It queries whether there is a cost number in the historical medical data, that is, it can query the corresponding data field. When the corresponding "cost" field is queried, the data field corresponds to the data to be processed. The conversion logic corresponding to the cost data is to split the cost into detailed costs. According to the conversion logic, the cost data is split to obtain the detailed costs as hospitalization costs, prescription costs, outpatient costs, etc., thereby completing the standardized processing of the historical medical data to obtain standardized historical medical data.

[0080] In this embodiment, the acquired historical medical data can be standardized to obtain standardized historical medical data, thereby avoiding the inconsistency of data that leads to the inability to make predictions, and then the need to process the data one by one, which can improve the prediction efficiency and the prediction accuracy.

[0081] In one of the embodiments, predicted historical medical data is input into a prediction model, a prediction result corresponding to the medical data to be predicted is obtained according to a prediction method, and the prediction result is output, including: obtaining the medical data to be predicted whose cost exceeds a preset value in the prediction result; obtaining the identity information corresponding to the medical data to be predicted whose cost exceeds the preset value; and monitoring the medical behavior corresponding to the identity information.

[0082] Specifically, the identity information refers to the identity data of the user contained in the medical data to be predicted, and the identity information may be the user's name, age, contact information, ID card information, etc. When the medical data to be predicted received by the prediction platform is predicted cost data, the corresponding medical behavior can be monitored based on the predicted result of the predicted cost data. When the medical data to be predicted received by the prediction platform is predicted corresponding cost data, the cost data is predicted and the corresponding prediction result is output, and the prediction result is the corresponding predicted cost data. The obtained predicted cost data is compared with the preset value. When the predicted cost is higher than the preset value, the cost displayed by the prediction result will have the risk of improper medical treatment, such as indiscriminate prescription of drugs, etc., and corresponding medical behavior is required, that is, the prediction platform obtains the identity information corresponding to the medical data to be predicted that exceeds the preset value of the cost data, so as to monitor the medical behavior of the user corresponding to the identity information according to the identity information.

[0083] In this embodiment, after using the prediction model to predict the data to be predicted, the cost data exceeding the preset value is queried, and then the medical data to be predicted corresponding to the cost is queried, and the medical behavior of the identity information corresponding to the medical data to be predicted is monitored. The corresponding users can be monitored to enhance the scope of application.

[0084] It should be understood that although Figure 2-3 The steps in the flowchart are shown in sequence as indicated by the arrows, but these steps are not necessarily executed in the order indicated by the arrows. Unless otherwise specified in this document, there is no strict order restriction for the execution of these steps, and these steps can be executed in other orders. Moreover, Figure 2-3 At least part of the steps may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily executed at the same time, but can be executed at different times. The execution order of these sub-steps or stages is not necessarily sequential, but can be executed in turn or alternately with other steps or at least part of the sub-steps or stages of other steps.

[0085] In one embodiment, Figure 4 As shown, a medical data prediction device 400 is provided, comprising: an acquisition module 410, a historical medical data processing module 420, a prediction parameter processing module 430 and a prediction module 440, wherein:

[0086] The acquisition module is used to acquire the medical data to be predicted and the historical medical data corresponding to the medical data to be predicted, and select the prediction method corresponding to the medical data to be predicted.

[0087] The historical medical data processing module is used to standardize the historical medical data to obtain standardized historical medical data.

[0088] The prediction parameter processing module is used to process the prediction parameters in the standardized historical medical data according to the prediction method to obtain the predicted historical medical data.

[0089] The prediction module is used to input the predicted historical medical data into the prediction model, obtain the prediction results corresponding to the medical data to be predicted according to the prediction method, and output the prediction results. The prediction model is used to characterize the corresponding relationship between the predicted historical medical data and the prediction results.

[0090] In one embodiment, the prediction parameter processing module includes:

[0091] The query unit is used to query the region information associated with the medical data to be predicted according to the prediction method.

[0092] The selection unit is used to select discarded data from the standardized historical medical data according to regional information.

[0093] Add a unit for adding deprecation labels to deprecated data.

[0094] The extraction unit is used to extract the standardized historical medical data without the abandoned labels to obtain the predicted historical medical data.

[0095] In one embodiment, the medical data prediction device 400 may include:

[0096] The to-be-replaced data query module is used to query whether there is to-be-replaced data in the standardized historical data when the selected prediction method is the linear prediction method.

[0097] The first preset replacement value query module is used to query the first preset replacement value corresponding to the data to be replaced when there is data to be replaced.

[0098] The first replacement module is used to replace the data to be replaced with a first preset replacement value to obtain predicted historical medical data.

[0099] In one embodiment, the medical data prediction device 400 further includes:

[0100] The replacement range query module is used to query the replacement range corresponding to the data to be replaced when the first preset replacement value corresponding to the data to be replaced is not found.

[0101] The second replacement module is used to select a second preset replacement value corresponding to the replacement range, and use the second preset replacement value to replace the data to be replaced to obtain predicted historical medical data.

[0102] In one embodiment, the historical medical data processing module may include:

[0103] The to-be-processed data acquisition unit is used to acquire the to-be-processed data in the standardized historical medical data and extract the data fields corresponding to the to-be-processed data.

[0104] The conversion logic selection unit is used to select the conversion logic corresponding to the data field.

[0105] The conversion unit is used to convert the to-be-processed data of the corresponding field into standardized historical medical data according to the conversion logic.

[0106] In one embodiment, the medical data prediction device may include:

[0107] The acquisition module can also be used to obtain the medical data to be predicted whose cost exceeds the preset value in the prediction results.

[0108] The identity information acquisition module is used to obtain the identity information corresponding to the medical data to be predicted whose cost exceeds a preset value.

[0109] The monitoring module is used to monitor the medical behavior corresponding to the identity information.

[0110] For the specific definition of the medical data prediction device, please refer to the definition of the medical data prediction method above, which will not be repeated here. Each module in the above medical data prediction device can be implemented in whole or in part by software, hardware and a combination thereof. The above modules can be embedded in or independent of the processor in the computer device in the form of hardware, or can be stored in the memory of the computer device in the form of software, so that the processor can call and execute the operations corresponding to the above modules.

[0111] In one embodiment, a computer device is provided. The computer device may be a server, and its internal structure diagram may be as follows: Figure 5As shown. The computer device includes a processor, a memory, a network interface and a database connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store medical data prediction data. The network interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, a medical data prediction method is implemented.

[0112] Those skilled in the art will understand that Figure 5 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0113] In one embodiment, a computer device is provided, including a memory and a processor, wherein the memory stores a computer program, and the processor implements the following steps when executing the computer program: obtaining medical data to be predicted and historical medical data corresponding to the medical data to be predicted, and selecting a prediction method corresponding to the medical data to be predicted. Standardizing the historical medical data to obtain standardized historical medical data. According to the prediction method, processing the prediction parameters in the standardized historical medical data to obtain predicted historical medical data. Inputting the predicted historical medical data into a prediction model, obtaining a prediction result corresponding to the medical data to be predicted according to the prediction method, and outputting the prediction result, wherein the prediction model is used to characterize the corresponding relationship between the predicted historical medical data and the prediction result.

[0114] In one embodiment, when the processor executes the computer program, the prediction parameters in the standardized historical medical data are processed according to the prediction method to obtain the predicted historical medical data, which may include: querying the region information associated with the medical data to be predicted according to the prediction method. Selecting the abandoned data in the standardized historical medical data according to the region information. Adding an abandoned label to the abandoned data. Extracting the standardized historical medical data without the abandoned label to obtain the predicted historical medical data.

[0115] In one embodiment, when the processor executes the computer program, the following steps are further implemented: when the prediction method is selected as the linear prediction method, it is queried whether there is data to be replaced in the standardized historical data. When there is data to be replaced, it is queried for a first preset replacement value corresponding to the data to be replaced. The data to be replaced is replaced with the first preset replacement value to obtain predicted historical medical data.

[0116] In one embodiment, when the processor executes the computer program, the following steps are further implemented: when the first preset replacement value corresponding to the data to be replaced is not found, the replacement range corresponding to the data to be replaced is queried, a second preset replacement value corresponding to the replacement range is selected, and the data to be replaced is replaced with the second preset replacement value to obtain predicted historical medical data.

[0117] In one embodiment, when the processor executes the computer program, the historical medical data is standardized to obtain the standardized historical medical data, including: obtaining the data to be processed in the standardized historical medical data, and extracting the data field corresponding to the data to be processed. Selecting the conversion logic corresponding to the data field. According to the conversion logic, the data to be processed of the corresponding field is converted to obtain the standardized historical medical data.

[0118] In one embodiment, when the processor executes the computer program, the following steps are further implemented: obtaining medical data to be predicted whose cost exceeds a preset value in the prediction result; obtaining identity information corresponding to the medical data to be predicted whose cost exceeds the preset value; and monitoring the medical behavior corresponding to the identity information.

[0119] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the following steps are implemented: obtaining medical data to be predicted and historical medical data corresponding to the medical data to be predicted, and selecting a prediction method corresponding to the medical data to be predicted. The historical medical data is standardized to obtain standardized historical medical data. According to the prediction method, the prediction parameters in the standardized historical medical data are processed to obtain predicted historical medical data. The predicted historical medical data is input into a prediction model, and a prediction result corresponding to the medical data to be predicted is obtained according to the prediction method, and the prediction result is output. The prediction model is used to characterize the corresponding relationship between the predicted historical medical data and the prediction result.

[0120] In one embodiment, when the computer program is executed by the processor, the prediction parameters in the standardized historical medical data are processed according to the prediction method to obtain the predicted historical medical data, which may include: querying the region information associated with the medical data to be predicted according to the prediction method. Selecting the discarded data in the standardized historical medical data according to the region information. Adding a discarded label to the discarded data. Extracting the standardized historical medical data without the discarded label to obtain the predicted historical medical data.

[0121] In one embodiment, when the computer program is executed by the processor, the following steps are further implemented: when the prediction method is selected as the linear prediction method, it is queried whether there is data to be replaced in the standardized historical data. When there is data to be replaced, the first preset replacement value corresponding to the data to be replaced is queried. The data to be replaced is replaced with the first preset replacement value to obtain the predicted historical medical data.

[0122] In one embodiment, when the computer program is executed by the processor, the following steps are further implemented: when the first preset replacement value corresponding to the data to be replaced is not found, the replacement range corresponding to the data to be replaced is queried, a second preset replacement value corresponding to the replacement range is selected, and the data to be replaced is replaced with the second preset replacement value to obtain predicted historical medical data.

[0123] In one embodiment, when the computer program is executed by a processor, the historical medical data is standardized to obtain standardized historical medical data, including: obtaining the data to be processed in the standardized historical medical data, and extracting the data field corresponding to the data to be processed. Selecting the conversion logic corresponding to the data field. According to the conversion logic, the data to be processed of the corresponding field is converted to obtain standardized historical medical data.

[0124] In one embodiment, when the computer program is executed by the processor, the following steps are further implemented: obtaining medical data to be predicted whose cost exceeds a preset value in the prediction result; obtaining identity information corresponding to the medical data to be predicted whose cost exceeds the preset value; and monitoring the medical behavior corresponding to the identity information.

[0125] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

[0126] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0127] The above-mentioned embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the invention patent. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the patent of the present application shall be subject to the attached claims.

Claims

1. A medical data prediction method, the method comprising: Acquire medical data to be predicted and historical medical data corresponding to the medical data to be predicted, and select a prediction method corresponding to the medical data to be predicted; the medical data to be predicted is used to represent medical behavior data that can be predicted within a preset time period; Standardizing the historical medical data to obtain standardized historical medical data; According to the prediction method, querying the region information associated with the medical data to be predicted; Selecting discarded data from the standardized historical medical data according to the regional information; Adding a deprecated label to the deprecated data; Extracting the standardized historical medical data without the abandoned label to obtain predicted historical medical data; Inputting the predicted historical medical data into a prediction model, obtaining a prediction result corresponding to the medical data to be predicted according to the prediction method, and outputting the prediction result, wherein the prediction model is used to characterize the corresponding relationship between the predicted historical medical data and the prediction result; The method further comprises: When the prediction method is selected as a linear prediction method, querying whether there is data to be replaced in the standardized historical data; When there is data to be replaced, querying a first preset replacement value corresponding to the data to be replaced; Replacing the data to be replaced with the first preset replacement value to obtain predicted historical medical data; The step of inputting the predicted historical medical data into a prediction model, obtaining a prediction result corresponding to the medical data to be predicted according to the prediction method, and outputting the prediction result includes: Acquire the medical data to be predicted whose cost exceeds a preset value in the prediction result; Obtaining identity information corresponding to the medical data to be predicted whose cost exceeds a preset value; The medical behavior corresponding to the identity information is monitored.

2. The method according to claim 1, characterized in that After querying the first preset replacement data value corresponding to the data to be replaced, the method further includes: When the first preset replacement value corresponding to the data to be replaced is not found, then querying the replacement range corresponding to the data to be replaced; A second preset replacement value corresponding to the replacement range is selected, and the data to be replaced is replaced with the second preset replacement value to obtain predicted historical medical data.

3. The method according to claim 1, characterized in that The step of standardizing the historical medical data to obtain standardized historical medical data includes: Acquire the data to be processed in the standardized historical medical data, and extract the data fields corresponding to the data to be processed; Selecting conversion logic corresponding to the data field; According to the conversion logic, the data to be processed of the corresponding field is converted into standardized historical medical data.

4. A medical data prediction device, characterized in that: The device comprises: An acquisition module, used to acquire the medical data to be predicted and the historical medical data corresponding to the medical data to be predicted, and select a prediction method corresponding to the medical data to be predicted; the medical data to be predicted is used to represent the medical behavior data that can be predicted within a preset time period; A historical medical data processing module, used for standardizing the historical medical data to obtain standardized historical medical data; A prediction parameter processing module, used for querying the region information associated with the medical data to be predicted according to the prediction method; selecting the discarded data in the standardized historical medical data according to the region information; adding a discarded label to the discarded data; extracting the standardized historical medical data without the discarded label to obtain the predicted historical medical data; A prediction module, used for inputting the predicted historical medical data into a prediction model, obtaining a prediction result corresponding to the medical data to be predicted according to the prediction method, and outputting the prediction result, wherein the prediction model is used for characterizing the corresponding relationship between the predicted historical medical data and the prediction result; The device also includes: A to-be-replaced data query module, used for querying whether there is to-be-replaced data in the standardized historical data when the prediction mode is selected as a linear prediction mode; A first preset replacement value query module, used for querying a first preset replacement value corresponding to the data to be replaced when there is data to be replaced; A first replacement module, using the first preset replacement value to replace the data to be replaced to obtain predicted historical medical data; The acquisition module is further used to acquire the medical data to be predicted whose cost exceeds a preset value in the prediction result; An identity information acquisition module, used to acquire the identity information corresponding to the medical data to be predicted whose cost exceeds a preset value; A monitoring module is used to monitor the medical behavior corresponding to the identity information.

5. The device according to claim 4, characterized in that The device also includes: A replacement range query module, configured to query a replacement range corresponding to the data to be replaced when a first preset replacement value corresponding to the data to be replaced is not found; The second replacement module is used to select a second preset replacement value corresponding to the replacement range, and use the second preset replacement value to replace the data to be replaced to obtain predicted historical medical data.

6. The device according to claim 4, characterized in that The historical medical data processing module includes: A to-be-processed data acquisition unit, used to acquire the to-be-processed data in the standardized historical medical data, and extract data fields corresponding to the to-be-processed data; A conversion logic selection unit, used to select a conversion logic corresponding to the data field; A conversion unit is used to convert the to-be-processed data of the corresponding field according to the conversion logic to obtain standardized historical medical data.

7. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 3 are implemented.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 3 are implemented.